Skip to main content

Full text of "What Was Artificial Intelligence?"

See other formats




Sue Curry Jansen 


WHAT WAS ARTIFICIAL 
INTELLIGENCE? 


MEDIASTUDIES.PRESS 


What Was Artificial Intelligence? includes an eponymous chapter originally published as Sue Curry Jansen, 
“What Was Artificial Intelligence?” in Critical Communication Theory: Power, Media, Gender, and Technology 
(Lanham, MD: Rowman Littlefield, 2002). Reprinted with permission. 

Published by MEDIASTUDIES.PRESS in the MEDIA MANIFOLD series 

mediastudies.press | 414 W. Broad St., Bethlehem, PA 18018, USA 

New materials are licensed under a Creative Commons Attribution-Noncommercial 4.0 (CC BY-NC 4.0) 
COVER DESIGN: Jeff Pooley 

COPY EDITING: Emily Alexander 

CREDIT FOR COVER: Massimo Botturi on Unsplash 

CREDIT FOR LATEX TEMPLATE: Book design inspired by Edward Tufte, by The Tufte-LaTeX Developers 
ISBN 978-1-951399-06-1 (print) | 978-1-951399-05-46 (ebook) 

DOI 10.32376/3f8575cb.0cc62523 


LIBRARY OF CONGRESS CONTROL NUMBER 2022930299 


Edition 1 published in April 2022 


Contents 


INTRODUCTION 1 


WHAT WAS ARTIFICIAL INTELLIGENCE? 11 


Introduction 


[Artificial intelligence is] the conjecture every aspect of learning or any 
other feature of intelligence can in principle be so precisely described 
that a machine can be made to simulate it. 


—John McCarthy and Marvin Minsky, 1955" 


[AI is] the scientific understanding of the mechanisms underlying 
thought and intelligent behavior and their embodiment in machines. 


—Association for the Advancement 
of Artificial Intelligence, 20227 


Al is the ability of a machine to display human-like capabilities such as 
reasoning, learning, planning and creativity. 


—European Parliament, 2020 


ARTIFICIAL INTELLIGENCE (AI) advocates generally describe it in 
the future tense. By inverting that convention in my 2002 essay, I in- 
tended to signal that AI also had a past—by then a half century—long 
history of extravagant forecasting—which was overdue for critical 
examination. I also wanted to suggest that, at that point, Al's fu- 
ture was uncertain, as the field was undergoing a period of critical 
reassessment. 

Almost from artificial intelligence’s inception in the 1950s, AI re- 
searchers had been periodically announcing that they were on the 
threshold of revolutionary discoveries that would radically transform 
human life as we know it, in ways that we could not begin to grasp. 
Artificial intelligence would, we were told, create a form of super in- 
telligence many times greater than human intelligence, which would 
continue to perfect itself through machine learning, leaving us slow- 
witted humans behind. AI enthusiasts celebrated the prospect, seeing 
themselves as either creating or bearing witness to the next step in 


doi 


‘John McCarthy et al., “A Proposal 
for the Dartmouth Summer Research 
Project on Artificial Intelligence” 
(unpublished manuscript, August 

31, 1955). McCarthy and Minsky are 
normally credited with drafting the 
proposal, and McCarthy with coining 
the “artificial intelligence” phrase. 
?"Welcome to the Association for the 
Advancement of Artificial Intelligence!" 
Association for the Advancement of 
Artificial Intelligence . 


3 “What Is Artificial Intelligence and 
How Is It Used?” European Parliament 
News, September 4, 2020. 


2 WHAT WAS ARTIFICIAL INTELLIGENCE? 


evolution. Sci-fi narratives multiplied and greatly amplified AI futur- 
ism, whether as deliverance or as the impending doom of a robotic 
apocalypse. 

Fact or fiction, good or evil: The audacity of the message seemed 
to infuse AI with independent agency, a godlike mind and destiny of 
its own. This deflected attention from the actual networks of military, 
political, and economic interests promoting its development. It also 
absolved the hubris of the scientists attending the nativity of the 
electronic marvel, because they seemed to cast themselves as mere 
messengers or apostles serving a higher power. 

By the 1990s, however, the air was growing thin. A half century 
of extravagant promises, substantial public investments, and meager 
visible returns led to disenchantment with AI’s top-down research 
paradigm. This early work drew on traditional philosophical studies 
of logic and reasoning processes to develop its models, with aspira- 
tions to formalize common-sense reasoning processes.* For AI, it was 
a time for rethinking and retooling, which suggested that AI’s hold 
on the future tense was, at best, tenuous. 

It was an opportune moment to rethink my own work on AI as 
well, which I had first undertaken in the mid-1980s, when enthusi- 
asm for the promise of artificial intelligence was at a peak. The 2002 
essay was not, however, intended as an obituary for artificial intelli- 
gence. No one expected its advocates to shut down shop. But at the 
time it did seem that the transcendent vision of the first generation 
of AI researchers—what is now sometimes referred to as “strong” 
artificial intelligence—was undergoing a radical deflation, and that it 
might be prudently retired. 

The results of my millennial reassessment of AI, “What Was Arti- 
ficial Intelligence?” is reproduced here in its original form, without 
any emendations.° It is not a history of the science or mathematics 
of AI. That lies far beyond my competence. Rather, it is an account of 
the stories the AI scientists have told themselves, each other, and the 
world about the form of intelligence they hoped to create. In telling 
those stories, they also tell us a great deal about themselves. 


ARTIFICIAL INTELLIGENCE PARATEXTS AND 
Hyper Cycles 


In literary theory, a paratext is “a text that relates to (or mediates) 
another text (the main work) in a way that enables the work to be 
complete and to be offered to its readers and, more generally, to the 
public.” It has been described metaphorically as a “threshold” or 
“vestibule” which allows readers to enter a text.° 


+Selmer Bringsjord and Naveen Sundar 
Govindarajulu, “Artificial Intelligence,” 
in The Stanford Encyclopedia of Philos- 
ophy, ed. Edward N. Zalta, accessed 
August 16, 2020 . 


5 The essay originally appeared as Sue 
Curry Jansen, “What Was Artificial 
Intelligence?” in Critical Communica- 
tion Theory: Power, Media, Gender, and 
Technology (Lanham, MD: Rowman & 
Littlefield, 2002), 123-54. Reprinted by 
permission. 


® Roswitha Skare, "Paratexts," in Ency- 
clopedia of Knowledge Organization, ed. 

Bjorn Hammarfelt and Claudio Gnoli 
(International Society for Knowledge 

Organization). 


INTRODUCTION 


“What Was Artificial Intelligence?” offers a critical analysis of 
twentieth-century paratexts of the AI movement: the programmatic 
descriptions, manifestos, and interviews that AI scientists used to 
explain what they thought they were doing when they did their 
research. Parascientific texts are similar to corporate mission state- 
ments, which are designed to cultivate and promote positive re- 
sponses to an enterprise. The target audiences for the parascientific 
texts of AI are potential public and private funding agencies like 
the National Science Foundation or the Defense Advanced Research 
Projects Agency (DARPA), policy makers and administrators of uni- 
versities and scientific research institutes, as well as scientists in other 
areas of expertise, science buffs, and journalists. So these parasci- 
entific texts do not assume or require competency in the technical 
aspects of AI. They are, in effect, translations or narrative accounts of 
the techno-sciences. 

Artificial intelligence advocates have been especially prolific 
in their production of paratexts, perhaps because the technology 
they are seeking to develop is unprecedented, esoteric, and ethe- 
real, and its delivery date is indefinite (although almost always de- 
scribed as near). According to widely accepted market analytics, new 
technologies—those that make it to market—typically undergo “hype 
cycles” of initial over-enthusiasm, followed by disillusionment, and 
then a plateau of productivity as the product is realistically assessed 


3 


and its utility demonstrated in the marketplace.” 7 Jackie Fenn, “Understanding Gartner’s 


It is generally agreed by AI chroniclers that the artificial intelli- face. 
gence movement has gone through two major hype cycles that have 

ended in disappointment. There is little agreement on exact dates or 

the specific AI visions that failed to meet expectations. There does, 

however, seem to be agreement that AI hype-cycle peaks are unusu- 

ally steep and its descents exceptionally low. In fact, the lows are so 

low that the AI community refers to them as AI “winters,” borrow- 

ing the seasonal trope from the Cold War concept of nuclear winter, 


when life on planet Earth is extinguished by nuclear devastation.® 5 Bret Kinsella, “Gartner Hype Cycle 
Suggests Another AI Winter Could be 


The disparity in dating AI low points—AI winters—is a function : 
Near,” Voicebot, November 5, 2017. 


of whether an analysis focuses primarily on (1) research funding, 

(2) breakthroughs or failures in specific Al-based technologies or 
promises, or (3) performance in the marketplace. While these dynam- 
ics are interrelated, their timing is sequential, which explains the dat- 
ing disparities. For our purposes, (1) funding is most relevant since 
most early AI research relied heavily on government funding, and its 
ebbs and flows have immediate impact on AI researchers. This dates 
the first AI winter to the early 1970s, when mechanical translation, 

a Cold War priority, was declared a hopeless failure and US govern- 
ment funding dried up. AI research in the UK also declined during 


Hype Cycles, 2007,” Gartner Research, 


4. WHAT WAS ARTIFICIAL INTELLIGENCE? 


the same approximate timeframe in response to disappointment in 
Al’s military applications.? 

By the end of the 1970s, however, there were buds on the AI tree 
again, in the US at least, with excitement about the potential of neu- 
ral networks: AI modeling based on biological, rather than logical, 
models of intelligence. Interest peaked in the mid-1980s, but Al’s 
second winter had set in by the early 1990s—and lasted so long that 
one observer referred to it as a “mini ice age instead of a winter.”*° 
Artificial intelligence research was not completely abandoned during 
AI winters, but funding was scarce and researchers tended to avoid 
the inflated “AI” label, adopting more modest descriptors such as 
expert systems, machine learning, informatics, pattern recognition, or 
knowledge-based systems."* 

The second winter lasted through the 2008 global financial cri- 
sis. The past decade has, however, inaugurated a new AI hype cy- 
cle with momentum that dwarfs the two previous cycles. Not only 
are governments heavily investing in AI for strategic and economic 
purposes, but there have also been large infusions of Silicon Valley 
wealth and other private investments in AI research and develop- 
ment. Even more significantly, the US and China have entered into a 
global competition for AI supremacy, comparable to the US-USSR’s 
Cold War space race.'* The World Economic Forum has also cast AI 
in a central role in what it is calling the “Fourth Industrial Revolu- 
tion.”*3 In the current hype cycle, “AI is the new gold.”"4 


FossiL POETRY 


My original engagement with AI evolved out of a long-standing in- 
terest in the sociology and politics of knowledge—in this instance, 
the human factors that shape scientific knowledge. In the mid-1980s, 
a number of path-breaking studies of gender and science were pub- 
lished. I followed that early literature closely until the floodgates it 
opened made it impossible for any single scholar to follow it all. 

At the time, I thought artificial intelligence could offer an espe- 
cially rich resource for studying the social constituents of science 
because of its future orientation, as well as its departure from the 
usual protocols of empirical science. That is, AI does not exist in na- 
ture. It can’t be apprehended by the senses, observed in the wild, or 
dissected in the laboratory. Rather, AI is a projection of the hopes and 
dreams of AI researchers. 

It is science in a formative stage. The development of artificial in- 
telligence can be studied in real time, unlike so many classic studies 
in the history and sociology of science, which reach into the past to 
expose the social factors in the development of science, usually fo- 


9 Gary Yang, “AI Winter and its 
Lessons,” in The History of Artificial 
Intelligence (University of Washington, 
2006). 


© Heikki Ailisto, “AI Winter is Coming 
Unless We Change Course,” VTT, July 
23, 2019. 


™ Yang, “AI Winter and its Lessons.” 


* Nicholas Thompson, “The AI Cold 
War that Threatens Us All,” Wired, 
October 23, 2018. 

3 Klaus Schwab, “The Fourth Industrial 
Revolution: What it Means, How to 
Respond,” World Economic Forum, 
January 14, 2016. 

“4 Andreas Triantafyllopoulos, “Why the 
Current AI Gold Rush Must Not Fail,” 
Towards Data Science, August 30, 2019. 


INTRODUCTION 


cusing on discredited or superseded scientific claims. So, artificial 
intelligence seemed an ideal case study for sociologists of knowledge. 

AI was also of interest from another perspective. For four decades, 
global politics had been organized around the generative metaphor 
of the Cold War. The 1980s were a transformative period, from glas- 
nost to the fall of the Berlin Wall and the demise of the Soviet Union. 
Change was in the air. Advances in computer technology, including 
the PC revolution, were also transforming business, as instant inter- 
national cash transfers became possible. Policy makers and pundits 
were vying to name this new constellation, to capture its gestalt. The 
“information age” and “information economy” were gaining traction. 
If AI could keep its promises, it would be the engine of the future. 
Because of its resonance within popular culture, references to AI, 
no matter how vague, seemed to add scientific authority to the ide- 
ological thrust of political and economic claims. So the paratexts of 
AI possessed significant geopolitical relevance in the 1980s, as well 
as sociological interest, even though “globalization” ultimately won 
the naming game. Since then, new efforts to fill the void with AI 
futurism have gained traction. 

To summarize, then, during my early studies of AI in the mid- 
1980s, AI was at the peak of its second hype cycle. My return to the 
topic in 2000 was at the low point in that cycle, Al’s second winter. 
Today we are at or near the pinnacle of a third cycle. The tenor of the 
parascientific texts of AI reflect these temporal locations: The hyper- 
bole soars approaching the peaks, and is chastened in the valleys. 

Until recently, however, there has been relatively little critical 
scholarly analysis of artificial intelligence as a social construct and 
political force. That is now changing, and changing rapidly and dra- 
matically. Writing in 2021, AI scholar Kate Crawford asserted, “A 
decade ago, the suggestion that there could be a problem of bias in 
artificial intelligence was unorthodox.”" ** Kate Crawford, Atlas of AI (New 

In 1986, when I first broached the subject, it was not just unortho- Haver sale: Univcraity, Eeeeenaee)> 
dox. To suggest that AI models might contain social fingerprints 
approached heresy. There were, however, a few prominent critics of 
Al's inflated claims, most notably computer scientist Joseph Weizen- 
baum and philosopher Hubert Dreyfus. Nevertheless, to suggest that 
AI was gendered was beyond the pale, although Weizenbaum seemed 
to intuit it in his reference to AI modelers as “big children” who have 


16 


not given up their “sandbox fantasies” or dreams of omnipotence. *© Joseph Weizenbaum, “Not Without 
8 p 


When he wrote this in 1988, there is no way that his metaphor would Rio 2 MRR aaHG JaRUTYy Lg88 ae: 
have evoked images of “big girl children” doing AI research, let alone 

indulging dreams of omnipotence. Elsewhere, Weizenbaum is un- 

ambiguous about the gender of the “unwashed and unshaven” who 


“are oblivious to their bodies and the world in which they move” and 


6 WHAT WAS ARTIFICIAL INTELLIGENCE? 


“exist, at least when so engaged, only through and for computers.”*7 
In a 2006 interview, he explicitly identified the masculinist bias of Al 
dreams of omnipotence and accused some AI extremists of “uterus 
envy.”18 

I published some of my early work on artificial intelligence and 
the information economy; however, these AI studies had almost no 
resonance. AI was not yet on the radar of social science research. It 
was only on my own radar because, at that time, my life was dis- 
proportionately populated by engineers, programmers, and hardline 
quantitative social scientists. My many part-time gigs as a gradu- 
ate student had included drawing computer flowcharts and editing 
field engineers’ reports. So I was often immersed in tech talk, with 
its instrumental values of parsimony, efficiency, and economy. To 
navigate this unfamiliar terrain, I drew on my sociological training 
and undertook an informal ethnographic study of the dialect of these 
tribes. It sensitized me to tech talk’s instrumental strengths as well 
as its blindspots. I also worked on survey research projects and took 
advanced courses in social science data analysis. As a result, I was 
keenly aware of the “extracting and abstracting” processes necessary 
to “clean up” survey research data in order to prepare it for com- 
puter processing. I had strong reservations about those efforts, as I 


suspected that some of the more revealing aspects of human behavior 


could be found in the anomalies that the clean-up scrubbed away. 


My interest in language had deeper roots. I learned early that sub- 


texts are often as important as texts, and sometimes more important. 
When I began studying AI, the linguistic turn in the social sciences 
was just beginning. So my own approach to discourse analysis in 
the 1970s and 1980s drew on an eclectic mix of sources drawn from 
social science and the humanities. Ralph Waldo Emerson’s descrip- 
tion of language as “fossil poetry” and his conception of metaphors 
as portals of knowledge also made a strong impression. In addition, 
I benefited from the wisdom of my dissertation advisor, Llewelyn 
Z. Gross (1914-2014), who studied “socio-logics”—the patterned 
reasoning processes in natural languages that formal logic does not 
accommodate—from philosophical and sociological perspectives.*? 
Later still, I discovered Lakoff and Johnson’s Metaphors We Live By 
(1980) and Philosophy in the Flesh (1999), which affirmed and legit- 
imized my intuitive sense that “by their metaphors you shall know 
them.”?° In the last thirty years there has been a revolution in the 
analysis of conceptual metaphors, based on Lakoff and Johnson’s 
work, which has transformed thinking about thinking and textual 
analysis in many fields. 

These were the sensibilities, along with feminist standpoint theory, 
that I brought to the analysis of the rhetoric of AI. Add the demo- 


7 Joseph Weizenbaum, Computer Power 
and Human Reason (San Francisco: W.H. 
Freeman and Company, 1976), 116. 


*8 Joseph Weizenbaum, interview 

by Gunna Wendt, in Islands in the 
Cyberstream: Seeking Havens of Reason in 
a Programmed Society, trans. Benjamin 
Fasting-Gray (Sacramento, CA: Litwin 
Books, 2015), 106. 


9 Llewelyn Z. Gross, “Intellectual 
Journey,” in Sociological Self-Images, ed. 
Irving Louis Horowitz (Beverly Hills, 
CA: Sage, 1969), 69-90. 


»° George Lakoff and Mark Johnson, 
Metaphors We Live By (Chicago: Univer- 
sity of Chicago Press, 1980); Lakoff and 
Johnson, Philosophy in the Flesh: The Em- 
bodied Mind and Its Challenge to Western 
Thought (New York: Basic Books, 1999). 


INTRODUCTION 


graphics of computer science in the 1980s. Unlike the earlier era, 
when women programmers played key roles in developing the field, 
computer science had become a boys’ club that required round-the- 
clock devotion in the elite graduate centers—what is now known as 
living on “Silicon time.” Epitomized by the legendary lore of MIT 
grad students’ nocturnal antics, it was common practice to restrict 
student access to scarce mainframe computer time to late-night hours 
at most research universities and, more often than not, the atmo- 
sphere became hostile to women. 

That, in summary, is the background to the essay: (1) the disparate 
timing of my two periods of AI inquiry—the first in the mid-1980s 
and the second in the early 2000s; (2) my perception of Al’s relevance 
to the sociologies of knowledge, politics, and gender; and (3) my ap- 
proach to the study of Al’s power-knowledge through the metaphors 
of its parascientific texts. 


SEASON THREE: RETURN OF THE ASTRONAUTS 


Most researchers revisit their early work with trepidation. We fre- 
quently encounter the voices of our former selves as alien, shudder 
at our stale epiphanies, and find perverse comfort in our low read- 
ership numbers on Google Scholar. Nonetheless, I went there again 
in 2021 because references to artificial intelligence suddenly seemed 
to be everywhere—not just in AI paratexts or science fiction—but in 
news stories, in international affairs reporting, in radio and television 
commercials, and in popular culture more generally. It seemed that 
globalization had been superseded by the AI gold rush. 

The mainstreaming of a new, globalized artificial intelligence hype 
cycle was clearly underway. I assumed AI had finally crossed the 
long-awaited threshold and had discredited my early skepticism. I 
took a deep dive into current AI paratexts to catch up on these new 
developments and, unexpectantly, found myself in familiar territory: 
similar tropes, the same breathless expectancy, even more extravagant 
sandbox fantasies. The new buzz is big data—your data, reader, 
extracted from your online activity. The AI component is pattern 
recognition, which abstracts and sorts that data. The methods are 
statistics and probability. 

Nonetheless, there has been substantial AI progress since my last 
visit. Much of that progress centers around developments in what is 
now known as “weak” or “narrow” AI, which can do specific tasks 
with far more speed, volume, and efficiency than any human. For 
example, Amazon’s recommendations to users based on their past 
searches and purchases. There have also been major victories for 
“strong” AI, too. In 1997, IBM’s Deep Blue computer famously beat 


8 WHAT WAS ARTIFICIAL INTELLIGENCE? 


world chess champion Gary Kasparov. In 2017, a computer beat the 
top-ranked player in the game of Go, which is considered infinitely 
more mathematically complex than chess; it also requires strategizing 
and trial-and-error machine learning. These AI developments have 
applications that extend far beyond games. According to artificial in- 
telligence enthusiasts, they move AI closer to passing the Turing Test, 
the standard for the achievement of artificial intelligence (thinking 
machines) set in 1950—the point when machines exhibit intelligi- 

ble behavior that is indistinguishable from human behavior. Some 
predict that this AI threshold will be crossed before the end of this 
decade. Others remain resolute in their skepticism. 

The emergence of AI skeptics is the real news—and from my per- 
spective, the good news for humankind. If, as Crawford claims, 
criticism of AI bias was unorthodox a decade ago, it is becoming 
increasingly mainstream today. When I wrote “What Was Artificial 
Intelligence?” I had Weizenbaum, Dreyfus, and computer scientist 
Bill Joy to lean on.?? Today there is a significant community of critical 
scholars studying many aspects of the artificial intelligence move- 
ment. 

In a 2000 jeremiad in Wired magazine (cited more fully in my es- 
say), Joy, co-founder and chief scientist of Sun Microsystems, warned 
that the technologies being developed in the twenty-first century— 
genetics, nanotechnology, and robotics—would be so powerful, ac- 
cessible, and amenable to abuse that they could pose a greater threat 
to humankind than the weapons of mass destruction of the twentieth 
century. He saw the astronautic fantasies of late twentieth-century Al 
scientists, which called for abandoning an over-populated, contami- 
nated, and warming Earth, in favor of interplanetary colonization— 
or, alternatively, merging with and becoming robots—as forms of 
denial that abdicate responsibility for life on earth. Joy pushed the 
panic button in hopes of initiating public dialogue about techno- 
futures which, to that point, had been shaped without it, by military 
strategists, military contractors, scientists, engineers, and by his fel- 
low tech entrepreneurs. 

That dialogue now exists and it is not confined to academic con- 
ferences, seminars, and computer labs, although robust research 
agendas are underway in those venues—far too robust to explore 
here, unfortunately. There is also a vibrant cyber-activist community 
responding to the surveillance regimes and authoritarianism enabled 
by digital technologies, working to create more just, equitable, trans- 
parent, and accountable forms of digital democracy. International 
critique and regulation of big tech companies is already well ad- 
vanced; and the United Nations has made combating gender bias in 
Ala priority, noting that research has “unambiguously” found gen- 


1 All of these sources are discussed 
and cited in “What Was Artificial 
Intelligence?” 


der biases in AI training data sets, algorithms, and devices.?? Similar 
racial biases have also been well-documented, including in devices 
already deployed in policing and criminal justice contexts. 

There are, however, very powerful forces aligned against these 
critical communities, with very deep pockets to fund political cam- 
paigns, lobbying, advertising, and public relations, to keep the cur- 
rent AI hype cycle spinning. The twenty-first-century titans of tech 
are the spiritual grandsons of the big-children-with-a-sandbox fan- 
tasies that Weizenbaum described. Unlike their forebears, however, 
they have the resources to indulge their astronautical fantasies of 
omnipotence and immortality. They are creating their own space 
programs, commissioning plans for colonizing the moon, and even 
exploring the possibility of producing an alternative universe, a vir- 
tual reality “metaverse,” where the humans left behind can go for 
fun and games when the physical world becomes too boring or un- 
pleasant. To wit, Jeff Bezos, Elon Musk, and Richard Branson have 
created their own space programs, and Bill Gates and Mark Zucker- 
berg are working on creating a “metaverse.” Some of these men also 
generously fund philanthropic initiatives, some to advance their own 
policy initiatives, but others presumably from altruistic motives. They 
have many sand pails, but in a democracy, policy is made by repre- 
sentatives of the public, in theory at least to serve the common good. 

Most of the new AI critics are not Luddites. They are not push- 
ing to abandon advanced computer research or to retire the robots. 
They are, however, calling for rejection of artificial intelligence’s post- 
human eschatology, and for replacing it with one that embraces and 
advances “[a] human form of life [that] is fragile, embodied in mortal 
flesh, time-limited, and irreproducible in silico.”*3 

To begin this reclamation of our tools and toys, artificial intel- 
ligence critic and policy expert Frank Pasquale calls for replacing 
the hype of AI with IA, “human intelligence augmentation.” He 
proposes a four-point set of “laws of robotics” to supersede science 
fiction writer Isaac Asimov’s 1942 laws for machines, developed in 
his short story, “Runaround”. 

They are: 


(i) “Robotic systems and AI should complement professionals, not 
replace them.” 

(ii) “Robotic systems and AI should not counterfeit humanity.” 

(iii) “Robotic systems and AI should not intensify zero-sum arms 
races.” 

(iv) “Robotic systems and AI must always indicate the identity of their 
creator(s), controller(s), and owner(s).”74 


INTRODUCTION 9 


> UNESCO, Artificial Intelligence and 
Gender Equality (Paris: UNESCO, 2020). 


23 Frank Pasquale, New Laws of Robotics 
(Cambridge, MA: Harvard University 
Press, 2020), 211. 


4 Pasquale, New Laws of Robotics, 3-14. 


10 WHAT WAS ARTIFICIAL INTELLIGENCE? 


Pasquale is writing from a different temporal and disciplinary lo- 
cation than my essay and he uses different words, but he draws the 
same conclusion as “What Was Artificial Intelligence?” 


CODA 


Rereading the 2002 essay, I did not blush. I found it timely, in places 
even eerily prescient. That is not a brag. It is a testament to the 
longevity of the hyperbolic mythopoetics of the artificial intelligence 
movement—from their embryonic inception in Turing’s 1950 essay, 
to their embellishment by the self-proclaimed descendants of Golem, 
and their inheritance by the present masters of the digital universe. 
While the essay is of some historical interest, it remains a relevant 
brief for the kind of humane and inclusive IA that Pasquale and 
many other critical AI scholars are now seeking. 


What Was Artificial Intelligence? 


We are all astronauts in this technological age, but the astronautic body 
of technological functioning there on the launch pad prepared and 
ready to depart the earth is a masculine figure. And the shadow of 

the abandoned body, the body left behind, exiled, imprisoned, and 
enchained, is the figure of the woman. 


—Robert D. Romanyshyn, Technology as Symptom and Dream, 1989" 


PROMETHEUS REBOUND: EVOLVING MODELS 
OF MIND 


THE STRONG research program for developing artificial intelligence 
was a Cold War ideological formation. Describing the artificial in- 
telligence movement in the past tense is an ironic reversal since it 
always described itself in the future tense. It never fully existed in the 
present. It was always becoming: its success forever contingent on a 
next step, a discovery that was just across the frontier of knowledge. 

The artificial intelligence movement (AIM) emerged as an identifi- 
able, if not yet organized, approach in the early 1950s. The computer, 
a technology with a long prehistory, became a reality at the end of 
World War II. With the subsequent invention of the transistor in 1949, 
the stage was set for the computer to become the defining technology 
of the late twentieth century. 

Germinal ideas for artificial intelligence (Al) can be traced to sep- 
arate but related attempts by Pascal and Leibniz to build machines 
that could calculate and thereby simulate functions of the human 
mind. The modern conception of artificial intelligence entered into 
the discourse of computer science with the 1950 publication of Alan 
Turing’s manifesto, “Computing Machinery and Intelligence,” which 
outlined a plan for creating computers that could think: a feat that 
Turing predicted would be achieved by the year 2000.7 A two-month- 
long conference of leading computer scientists in 1956, the Dart- 
mouth Summer Research Project on Artificial Intelligence, marked 


doi | This chapter is a reprint of Sue 
Curry Jansen, “What Was Artificial 
Intelligence?” in Critical Communica- 
tion Theory: Power, Media, Gender, and 
Technology (Lanham, MD: Rowman & 
Littlefield, 2002). The text appears as 
originally published, with the excep- 
tion of reference formatting and minor 
corrections. Reprinted by permission. 


* Robert Romanyshyn, Technology 
as Symptom and Dream (New York: 
Routledge, 2003), 171. 


? Alan M. Turing, “Computing Machin- 
ery and Intelligence,” Mind 59, no. 236 
(1950): 455- 


12 WHAT WAS ARTIFICIAL INTELLIGENCE? 


the formal emergence of artificial intelligence research as a “move- 
ment.”3 

Immediate inspiration for the project was drawn from the work 
of four generative thinkers of early computing. In addition to Tur- 
ing, whose 1936 paper “On Computable Numbers” described the 
theory of, specifications for, and limitations of “logic machines,”4 
they included John von Neumann, who headed the research team 
that designed and developed the modern, memory-based computer 
central processing unit (CPU), the computer architecture that is still 
used today; Norbert Wiener, who envisioned a new science of “cyber- 
netics”; and Claude Shannon, who developed information theory and 
inspired early interest in the social scientific study of communication. 
The researchers who actually formed and led the movement to de- 
velop Al included, among others, Alan Newell and Herbert Simon, 
both jointly affiliated with the RAND Corporation and Carnegie 
Institute of Technology (now Carnegie Mellon University); Marvin 
Minsky and Seymour Papert of Massachusetts Institute of Technol- 
ogy; and John McCarthy of Dartmouth and later Stanford. Most 
contemporary AI scientists have studied with one or more of these 
pioneers. 

What brought these men together was a common commitment to 
move beyond the then-prevalent understanding of computers as mere 
tools, advanced adding machines which could only do what they 
were told to do. The dominant view of the time was expressed in the 
familiar programmer’s motto: “garbage in, garbage out.” The goal 
of the AIM was to create computers that could “think” and learn. 

As Simon put it, the statement that “computers can do only what 
they are programmed to do is intuitively obvious, indubitably true, 
and supports none of the implications that are commonly drawn 
from it.”> AIM sought to create programs that would simulate the 
complexity of the human mind. These simulations would, however, 
amplify human reasoning powers, and would ultimately be more 
powerful than any single human mind: “It will be a program that 
analyzes, by some means, its own performance, diagnoses its failures, 
and makes changes that enhance its future effectiveness.”° 

AIM is now a half century old. Some former enthusiasts speak 
of it in the past tense: as a self-correcting intellectual movement 
that has transcended itself by serving as a launchpad to other en- 
deavors.” Others conceive of the movement as ongoing, but view its 
history as made up of two distinctive periods. Various pairings of 
adjectives have been used to define the shift in emphasis. The first 


Wu 


period, in which the “strong,” “top-down,” or “traditional” approach 
envisioned by Simon and Newell held sway, extended from the in- 


ception of the field to the mid-1980s. The second period of “weak,” 


3 The artificial intelligence movement 
has been widely chronicled. The brief 
overview of its history provided here 
draws primarily on the following 
sources: David Bolter, Turing’s Man: 
Westem Culture in the Computer Age 
(Chapel Hill: University of North 
Carolina Press, 1984); John L. Casti, 
Paradigms Regained (New York: Harper- 
Collins, 2000); Douglas R. Hofstadter, 
Gédel, Escher, Bach: An Eternal Golden 
Braid (New York: Random House, 
1979); George Johnson, Machinery of the 
Mind: Inside the New Science of Artificial 
Intelligence (New York: Random House, 
1986); and Sherry Turkle, The Second 
Self: Computers and the Human Spirit 
(New York: Simon & Schuster, 1984). 

4 Alan M. Turing, “On Computable 
Numbers, with an Application to the 
Entscheidungsproblem,” Journal of Math 


58 (1936). 


5 Johnson, Machinery of the Mind, 37. 


® Johnson, Machinery of the Mind, 37. 


7 Sherry Turkle, who has extensively 
chronicled the computer culture at 
MIT, reports, “Mainstream computer 
researchers no longer aspire to program 
intelligence into computers but expect 
intelligence to emerge from the interac- 
tions of small sub programs.” That is, 
scientists are no longer seeking to 


WHAT WAS ARTIFICIAL INTELLIGENCE? 13 


Wu 


“bottoms-up,” “emergent,” or “new wave” approaches emerged in 
the mid-1980s. The bottoms-up approach does continue some of the 
research program launched by the older, top-down tradition—for 
example, developing expert systems, robotics, and other commercial 
applications of AI. What marks the new wave as distinctive, however, 
is the reconfiguration of the metaphoric definitions of the field of 
inquiry. Logico-mathematical models give way to or merge with bio- 
logical metaphors, as the research goal is reconceived as the creation 
of artificial life (“A-life”) rather than artificial intelligence.® 

The top-down approach focuses on patterns and rules operating at 
the high level, symbol-processing structures of the brain, while ignor- 
ing its lower-level physical processes. The bottoms-up approach was 
a reaction against the failure of the top-down approach to produce 
significant results after more than thirty years of research. Whereas 
the top-down approach ignored biology, the bottoms-up approach 
took the position that the physical structure of the brain may account 
for its cognitive capacities. The bottoms-up approach seeks to de- 
sign computing devices that mimic the structure of the brain’s neural 
networks: that is, devices modeled on child development which can 
observe and learn.? This approach is also known as “connectionism” 
and is encompassed under the broad umbrella of the “new sciences 
of complexity.”*° Some chroniclers of the history of AI see the publi- 
cation of Minsky’s The Society of Mind (1987) as the benchmark; some 
date it to a conference on A-Life in 1987."' Most regard it as a more 
gradual shift away from the original vision: an evolutionary shift 
rather than a revolutionary displacement of paradigms. 

While the top-down versus bottoms-up distinction is useful for 
explaining the internal history of AI, it is already in some ways an 
arcane and, in the current fast-paced environment of technological 
change, archaic distinction. The end of the Cold War triggered a re- 
structuring of big science that was far more rapid, pervasive, and, by 
its own measures, much more successful than even the movers and 
shakers of this transformation anticipated in their most optimistic 
projections. The new research and development model streamlined 
and mainstreamed the old defense model for research and develop- 
ment by creating new, comprehensive partnerships of government, 
university, and corporate research and development initiatives—a 
model the Japanese had pioneered, to the dismay of the US govern- 
ment, in the 1970s and 1980s. Computer science and technologies, 
genetics, and bioengineering have been the leading edges of this 
new technological initiative; and commercialization of these fields 
fueled the unparalleled growth of US stock markets during the 1990s. 
The infusion of corporate capital has produced rapid advances in 
computer networking, robotics, and nanotechnologies that have tran- 


produce AI but rather AL (artificial 
life). The A-Life movement builds on 
the work of emergent AI research: a 
tradition that had been abandoned in 
the 1960s, but was rejuvenated by the 
shift to the bottoms-up tradition of AI. 
See Sherry Turkle, Life on the Screen: 
Identity in the Age of the Internet (New 
York: Simon & Schuster, 1995), 20. 


81 use Casti’s shorthand terminology, 
top-down and bottoms-up, through- 
out rather than Turkle’s AI and A-Life 
distinction, even though I am tempted 
by the greater drama of Turkle’s terms. 
Casti’s terms are cleaner and, in the 
case of bottoms-up, more encompass- 
ing. See Casti, Paradigms Regained. See 
also Hofstadter, Gédel, Escher, Bach; and 
Johnson, Machinery of the Mind. 


° Casti, Paradigms Regained. 


*° Heinz Pagels, The Dreams of Reason: 
The Computer and the Rise of the Sciences 
of Complexity (New York: Simon & 
Schuster, 1988). 

™ Marvin Minsky, Society of Mind (New 
York: Simon & Schuster, 1987). See also 
Turkle, Life on the Screen. 


14. WHAT WAS ARTIFICIAL INTELLIGENCE? 


scended AI without leaving it behind. Some of the leading AI scien- 
tists and all of the sites that housed leading AI laboratories continue 
to be key players in the creation of the scientific and technological 
infrastructure of the information economy. It is not too much of a 
stretch to say that many of the technovisions that were incubated 

in AI laboratories have been mainstreamed into our brave new info 
world. Just a decade ago, the utopian and dystopic projections of Al 
manifestos seemed unbelievable, woolly-headed sci-fi fantasies. Now, 
we are building the global infrastructure that supports them. 

The rise and fall of the “strong” artificial intelligence program 
roughly parallels the duration of the academic careers of the found- 
ing generation, although Minsky, who was a graduate student when 
he participated in the formative Dartmouth summer project, serves 
as a bridging figure. Its rise and decline also appears to coincide 
with the influence of the unity of science movement, of which it 
was part. The funding and fate of top-down AI were closely tied to 
the duration of the Cold War, with the bottoms-up transitional pe- 
riod coinciding with the US government’s expansion of its defense 
funding priorities to include economic “competitiveness.” The com- 
petitiveness thrust created the preconditions for jump-starting the 
information economy by underwriting the so-called greening of ar- 
tificial intelligence: the period when entrepreneurial AI scientists 
began aggressively promoting the commercial applications of their 
work, sometimes to the dismay of more idealistic Al founders like 
John McCarthy.’ 


PURE SCIENCE AND THE COLD WAR 


Like virtually all university-based computer science research during 
the Cold War, AI research was funded by the Department of De- 
fense’s Advanced Research Projects Agency (ARPA). Therefore, it was 
a player in the arms race with the Soviet Union and later the “com- 
petitiveness” race with Japan. Taken at face value, it was not a very 
effective player. In fact, it might have been viewed as an academic 
boondoggle: a metaphoric equivalent of a $7,000 Pentagon hammer. 
But taking AIM at face value grossly underestimates its accomplish- 
ments. Scientists associated with the movement made definitive con- 
tributions to the development of robotics and expert systems, which 
have had significant military and commercial applications. Top-down 
AI was also a very successful learning experience that taught sci- 
entists a great deal about the complexity of the brain and thereby 
provided the impetus for the development of what would become a 
new branch of psychology: cognitive science. 


* Johnson, Machinery of the Mind. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 15 


Artificial intelligence was part cover story as well as an important 
part of the real story of the early development of computer science. 
It leveraged the spectacular successes of the generation of Turing, 
von Neumann, Wiener, and Shannon, whose work had been sup- 
ported by unlimited wartime resources, into an equally ambitious 
ongoing program for basic research in computer science. “Basic” and 
“pure” were crucial adjectives for naming and claiming significant 
degrees of intellectual autonomy for government-funded research 
during the Cold War. The terms referred to research that did not 
have immediately apparent instrumental applications: for example, 
developing a computerized chess game that was smart enough to 
defeat the world’s top chess champions and thereby pass the Turing 
test for intelligence. This relative non-instrumentality was, of course, 
very instrumental to the research programs and careers of computer 
scientists. 

“Basic” still retains some of this patina in scientific grantsman- 
ship. “Pure” was, however, a crucial descriptor in the ideological 
and institutional struggles of the early Cold War period. Scientists, 
who chased defense dollars, professed their purity to try to fend 
off charges of scientific prostitution in the days when the govern- 
ment’s growing presence in the funding of private universities was 
unsettling to many in the academy. The unsettled ranged from tradi- 
tionalists, who wanted to preserve the relative insularity (cum purity) 
of the ivory tower, to liberals and leftists, critical of the Cold War pol- 
icy and of threats to intellectual freedom posed by what President 


Eisenhower called “the military-industrial complex.” 3 3 For a chronicle of how the Cold War 
climate impacted university life in a 
ae 5 : variety of disciplines, as seen from left 
practice free of social and political influences and interests. To the and liberal perspectives, see David 


contrary, I treat it as a strategic, ideological stance that artificial intel- Montgomery et al., The Cold War and 
The University: Toward an Intellectual 
History of the Postwar Years (New York: 
dual careers as defense researchers and academics. The ideology of New Press, 1997). 


When I use the term “pure” here, I am not suggesting a pristine 


ligence and other scientists used during the Cold War to justify their 


pure science also sometimes served as a temporary safe harbor dur- 
ing a politically complex and compromising era: the dark period of 
US government interrogations and purges of academics, intellectuals, 
writers, and other culture workers by the House Un-American Ac- 
tivities Committee and by Senator Joseph McCarthy. 

Scientists doing defense work were, by definition, always under in- 
tense scrutiny as potential national security risks. Much of their work 
was Classified and only accessible to those with government security 
clearances. In those days, the purer the science, the safer the scientist. 
I am not, however, suggesting that the scientists who embraced the 
sanctity of pure science were cynics, liars, propagandists, or scien- 
tific prostitutes, although some individual scientists may have been. 
Rather, I am saying that pure science was an ideal, an aspiration that 


16 WHAT WAS ARTIFICIAL INTELLIGENCE? 


had pragmatic as well as intellectual resonance. Like all potent ide- 
ological formations, it was a complex and fluid construct, part truth, 
part self-serving shield; it was a tool of power that could sometimes 
be used to hold the powerful accountable. 

The technovisions that now support the growth of the domestic US 
economy and its globalizing thrust are inspired to a significant de- 
gree by the achievements of the pure and impure sciences of AIM." 
Although some influential figures associated with AIM are now un- 
characteristically modest about their achievements, and, it would 
seem, almost eager to acknowledge their “failures,” the distance from 
the AI laboratories to the information economy is small. In fact, in 
some instances, it is just across the threshold: that much-vaunted 
next step. Both figuratively and literally, it is a step, sometimes di- 
rect, sometimes faltering, from publicly funded military research to 
publicly and privately supported applications of digital technologies, 
including the internet. “Convergence,” the hot techno buzzword of 
the 1990s, is being actualized in this century as a reengineering of 
society as well as technology.*> 

The significance of this reengineering is profoundly transforma- 
tive. An Wang, founder of Wang computers, maintained, “The dig- 
italization of information in all of its forms will probably be known 
as the most fascinating development of the twentieth century.”?° Ivan 
Illich underscored the revolutionary structural changes that digitiza- 
tion is bringing about. Conceiving of computers as agents of a new 
enclosure movement, he warns that computers “are doing to commu- 
nications what fences did to pastures and cars did to streets.”"7 In 
short, the digital revolution marks a deep structural shift in how we 
think, what we think about, how we communicate, how we relate to 
the material world and to one another, how we organize our work, 
and how we construct communities. 


CONTEXT AND LIMITS OF THIS CHAPTER 


The purpose of this chapter is not to assess the successes or failures 
of AI science as science. That is beyond my expertise. My goal is 
much more modest: to explore the rhetoric and mythopoetics of the 
parascientific discourse of artificial intelligence scientists. By parasci- 
entific discourse, I mean the programmatic descriptions, manifestos, 
and interviews that artificial intelligence scientists have used to ex- 
plain what they think they are doing when they do AI science. 

This paradiscourse might be conceived as functioning in academic 
science in the way that mission statements function in the corporate 
world. Both articulate the values, means, goals, and hopes of their 
enterprises. Like corporate mission statements, parascientific state- 


4 Many of the same cutting-edge AI 
scientists interviewed by Turkle and 
Grant Fjermedal in the 1980s are the 
same cutting-edge computer scientists 
that Bill Joy, a cutting-edge corporate 
scientist, interviewed for his current 
work on scientific futures. See Turkle, 
The Second Self; Grant Fjermedal, The 
Tomorrow Makers (New York: Macmil- 
lan, 1986); Bill Joy, “Why the Future 
Doesn’t Need Us,” Wired, April 2000. 


* Robert W. McChesney notes the 
migration of this term in Corporate 
Media and the Threat to Democracy (New 
York: Seven Stories, 1997). 


*© An Wang quoted in Tom Forester, 
High-Tech Society: The Story of the Infor- 
mation Technology Revolution (Oxford: 
Basil Blackwell, 1987), 1. 


7 Tvan Illich, “Silence is a Commons,” 
CoEvolution Quarterly 40 (1983): 5. The 
analogy to the enclosure movement 
that Illich embraces has had currency in 
communications since the 1980s. More 
recently, and apparently independently, 
the trope has gained currency among 
legal scholars. For a summary of this 
work, see James Boyle, “Fencing off 
Ideas: Enclosure and the Disappearance 
of the Public Domain,” Daedalus 131, 
no. 2 (2002). 


WHAT WAS ARTIFICIAL INTELLIGENCE? 17 


ments are, in a special sense, also public relations efforts. That is, 
they are purposively constructed to cultivate and promote positive 
perceptions; in the case of the parascientific discourse of Al, the in- 
tended audience appears to be other scientists, potential government 
and private sponsors, science buffs, and the general public. During 
the heyday of top-down Al, the forefathers and the founders of AI 
functioned as the practical philosophers of computer science; their 
influence was not limited to AI practitioners. 

The rhetoric of the parascientific discourses of the artificial in- 
telligence movement is remarkable on a number of counts. It does 
not use the flat, carefully measured language that experts on scien- 
tific writing recommend. To the contrary, it is frequently provocative 
and hyperbolic. Aphorisms, puns, and slogans are common, as are 
learned allusions to philosophy, literature, art, and music. Intrinsi- 
cally interesting numerical and visual puzzles and paradoxes are 
often used to illustrate points, and, I suspect, to engage and enter- 
tain readers who cannot fully follow or who might be bored by the 
accompanying technical explanations. Self-deprecating modesty and 
humor are sometimes deployed, but they are usually accompanied 
by dissembling winks. Expressions of self-doubt are, however, hard 
to find. Normally, authorial voices that aggressively flaunt their su- 
periority, even hubris, disturb and alienate readers; however, in the 
parascientific discourse of AI, this mode of address functions as a 
seductive hook. It uses inclusive pronouns and generous displays of 
encompassing “of course” constructions to flatter readers. It models 
readers as peers, colleagues, knowing and supportive companions; 
if readers take the hook, this mode of address seems to say that 
they too will be admitted to Mount Olympus where they will also 
see like gods and be like gods (or astronauts). Some AI spokesmen 
have spent most of their careers modeling natural language; they are 
acutely aware of how languages work, and how irony, poetry, and 
Aesopean indirection resist, mislead, and charm AI modeling at- 
tempts. And some of these men are very adept at using these tropes 
to engage readers. For example, Minsky’s discussion of metaphor 
is cutting-edge post-structuralism, but it has been cleanly shaven 
into clear, concise, and easily accessible prose.1® When these writ- 8 Minsky, Society of Mind. 
ers use synecdoche, they mean it: For Minsky, mind is a “society.” 
Top-downers are prone to what bottom-upper Douglas Hofstadter 
refers to as “ ‘Buck Rogers’ fantasies.”'? Some of these fantasies are 9 Hofstadter, Gédel, Escher, Bach, 601. 
presented in whimsically engaging prose, prefaced by almost child- 
like “what ifs.” Most of the writing is artful. A few authors need 
to be taken seriously as writers as well as thinkers: Hofstadter won 
a Pulitzer Prize for his remarkable book Gédel, Escher, Bach (1979). 
Clever, arrogant, self-serving, engaging, propagandistic, literate, 


18 WHAT WAS ARTIFICIAL INTELLIGENCE? 


playful, often facile, occasionally profound, sometimes outrageous, 
and usually interesting, the parascientific discourse of AIM inspires 
believers and incenses critics. 

While parascientific texts are clearly intended as a form of scien- 
tific outreach, frequently even proselytization, nonscientists are not 
encouraged to critically interrogate them. The late Isaac Asimov, who 
is regarded as the patron saint of “robotic ethics” by the AI com- 
munity, celebrated this resistance to external criticism of science in 
“Every Real Problem Can and Will Be Solved”: 


I’m a great one for iconoclasm. Given half a chance, I love to say some- 
thing shattering about some revered institution, and wax sarcastically 
cynical about Mother’s Day or apple pie or baseball. Naturally, though, 
I draw the line at having people say nasty things about institutions I 
personally revere. Like Science and Scientist, for instance (Capital S, 
you'll notice).?° 


Parascientific discourse is frequently treated with the same rever- 
ence as science. Where in traditional (preconstructivist) philosophies 
of science, the scientist is seen as a kind of miner who goes off and 
discovers precious ore, the parascientific writer, even the nonscien- 
tists among them, seem to see themselves as sharing the charisma 
of scientific discovery.*’ They go off and mine the texts and the talk 
of scientists, translate what they find into reader-friendly language, 
and then offer to share that precious metal with readers. Sociological 
analysis and rhetorical criticism seems to be all but proscribed.?? 

Turing himself imputed theological and philosophical significance 
to AI modeling; as a result, philosophers, unlike sociologists, have 
been part of AI paradiscourse almost since its inception. They have 
extensively criticized the ontological, logical, and linguistic assump- 
tions of the models of mind proposed by top-down AI scientists; see, 
for example, Dreyfus, Searle, Boden, Collins, and Pemose.?3 Else- 
where, I explored some of the ways that the logical structures of 
top-down AI models incorporate gendered assumptions.*4 See, for 
example, the comparisons and conflations of Maruyama’s concept of 
“classificatory information” and Gilligan’s typification of masculine 
modes of decision-making, or “morality of rules.” 

My purpose here is not to reiterate the philosophical critiques, 
but rather to use some of their scaffolding as support for investigat- 
ing some of the underexamined social constituents of the AI project. 
These constituents include the gendering of the language and as- 
sumptions of AI paradiscourse, and, to a lesser extent, the historically 
specific Cold War social formations of that gendered language—for 
example, doomsday thinking.*5 

AIM is an especially rich and unusually accessible site for excavat- 
ing the poetry in the paradigms of scientific thought. The nature of 


»° Isaac Asimov, “Every Real Problem 
Can and Will Be Solved,” in 2,500 Years 
of Science Writing, ed. Edmund Blarr 
Bolles (New York: W. H. Freeman, 


1999), 5-6. 


** George Johnson points out, “In our 
society, we make a distinction between 
the history of science and the history of 
everything else.” Within this proscrip- 
tion, only scientists are empowered to 
criticize science. Science writers, who 
assume the role of interpreters within 
the boundaries of this proscription, 
appear to share its protections. See 
Johnson, Fire in the Mind: Science, Faith, 
and the Search for Order (New York: 
Random House, 1996), 5. 

» For an example of this proscription 
at work, see the hierarchical framing 
of references to (e.g., “down among 
the sociologists”) and expressions 

of contempt for sociologists (Edin- 
burgh constructivists) who presume 

to question science, in Paul K. Gross 
and Norman Levitt, Higher Supersti- 
tion: The Academic Left and its Quarrels 
with Science (Baltimore: Johns Hopkins 
University Press, 1994). 


3 Hubert L. Dreyfus, What Computers 
Can't Do: A Critique of Artificial Reason 
(New York: Harper & Row, 1979); Hu- 
bert L. Dreyfus, What Computers Still 
Can't Do (Cambridge, MA: MIT Press, 
1992); John Searle, Intentionality: An 
Essay in the Philosophy of Mind (Cam- 
bridge: Cambridge University Press, 
1983); Margaret A. Boden, Computer 
Models of Mind (Cambridge: Cam- 
bridge University Press, 1988); and 
Harry M. Collins, Artificial Experts: 
Social Knowledge and Intelligent Machines 
(Cambridge, MA: MIT Press, 1993). 


WHAT WAS ARTIFICIAL INTELLIGENCE? 19 


the AI project itself, simulating or modeling minds, forced artificial 
intelligence scientists to consciously reflect upon the godlike roles 
they were playing in daring to try to create artificial life. It also re- 
quired them to carefully weigh the qualities of the human mind they 
wanted to incorporate in their models and the qualities they wanted 
to leave behind. 

AI science is unusual—an extreme case—in the history of Western 
science’s long struggle with dualism. As a mind modeling mind, the 
artificial intelligence scientist is both subject and object: the observer 
and the observed.?° He cannot deny his agency. Unlike other forms 
of scientific discourse, which attempt to erase all social fingerprints 
from scientific reason, artificial intelligence scientists recognize that 
their fingerprints are indelible. Some even seem to celebrate their 
presence: to engage in conscious myth-making about the significance 
of their work. For these reasons, the extreme case is also an ideal case 
for exploring the mythopoetics of scientific vision. 

AI scientists have spoken very freely and often quite extravagantly 
about their roles as modelers and about the qualities of their models. 
During World War II, women played substantial roles in wartime 
computing; however, nearly all artificial intelligence scientists during 
the Cold War era were men. They formed the so-called nerd or, until 
it became a pejorative term, hacker masculine subcultures of elite 
science and engineering schools; it is therefore not surprising that the 
AI manifesto writers are all male. 

Indeed the subworld of pre-PC academic and scientific computing 
was perhaps the purest post-World War II articulation of the monas- 
tic culture of science so painstakingly documented by David Noble 
in his underappreciated but groundbreaking contribution to both the 
history of science and the feminist critique of science, A World With- 
out Women: The Christian Clerical Culture of Western Science (1992).?7 
The nature of early computer technology and the rigid gender so- 
cialization of the Cold War era combined to make the subworlds of 
serious academic and scientific computing an exclusively male pre- 
serve. Use of mainframe computers was based on time-sharing. Typi- 
cally, by day the mainframe did the routine business of the university, 
its instructional and administrative tasks, and perhaps some of the 
work of senior researchers. By night, computer centers belonged to 
engineering and computer science graduate students, who basically 
lined up to run, then debug, then re-run the complex programs that 
demanded a lot of the computer’s time and memory. These gradu- 
ate students, who attended and taught classes by day, did their real 
work in and near the computer labs by night. Within the computer 
subculture, the mainframe was referred to as God, who determined 
the life (a successful run) or death (a glitch that needed debugging) 
of programs. 


>4 Sue Curry Jansen, “Is Information 
Gendered?” in Critical Communica- 
tion Theory: Power, Media, Gender, and 
Technology (Lanham, MD: Rowman 

& Littlefield, 2002). See also Jansen, 
“The Ghost in the Machine: Artificial 
Intelligence and Gendered Thought 
Patterns,” Resources for Feminist Re- 
search/Documentation sur la recherche 
feministe 17, no. 4 (1988): 4-7. 

5 Cynthia Enloe, The Morning After: 
Sexual Politics at the End of the Cold War 
(Berkeley: University of California 
Press, 1993). 

26 Hofstadter, Gédel, Escher, Bach. From 
the beginning, AI theorists were aware 
of this conundrum, but Hofstadter 
explores its implications with stunning 
virtuosity. 


27 David F. Nobel, A World Without 
Women: The Christian Clerical Culture of 
Western Science (New York: Alfred A. 
Knopf, 1992). 


20 WHAT WAS ARTIFICIAL INTELLIGENCE? 


The lumbering technology of the machines themselves demanded 
a kind of de facto near-equivalent of a vow of celibacy from their 
supplicants. They were expected to demonstrate their seriousness 
by periodically eating, sleeping, and socializing in the building that 
housed the computer. The overachievers, the nerds and hackers, vir- 
tually lived in and for the nighttime worlds of the labs. Like most 
all-male subcultures, this one had a dark underside in which male 
bonding was frequently mediated by shared misogynist and re- 
pressed homoerotic fantasies, jokes, and storytelling. Technology 
itself is sometimes eroticized within the subcultures of elite science, 


creating a kind of technoporn “that rouses prurient interest, demeans 


the powerless, eroticizes domination,” and sets up boundaries that 
signal they are off-limits to women and other outsiders.?8 

My analysis opens a rather narrow window onto that masculine 
subculture by exploring the mythopoetics of the technovisions of AI. 
It also examines the anxious image of masculinity that accompanies 
the generative metaphors that animate these visions. 


DREAMS OF REASON: WHEN DREAMERS DREAM 
THEY ARE DREAMING, ARE THEY AWAKE? 


Working largely independently of each other, constructivist and fem- 
inist analyses of science have exposed the fiction of “pure” science.?9 
They have established that science, like other noble and ignoble hu- 
man enterprises, is the work of mortal men and women, not of gods. 
Science is a social and cultural practice, which supports some of hu- 
mankind’s highest aspirations to and achievements of excellence. 
Until the twentieth century—and then only in atypical cases, for 
example Heisenberg’s physics, Gédel’s Theorem, and AI modeling— 
science has been a practice that has been secured in denial of its own 
nature.3° This denial has been deftly concealed and papered over for 
centuries by official histories and laundered origin stories. Anchored 
in the Western mind-body dualism, this denial makes doing science 
an “out-of-body experience.” The scientist seeks domination over 
nature by denying, implicitly or explicitly, that he is part of nature. 
His pretense to objectivity is maintained by detaching his mind from 
his body and the world, and by denying his mortality. This stance al- 
lows the scientist to believe that he is spying on the world from afar: 
viewing it dispassionately through God’s eyes or through the eyes 
of an astronaut. The gendering of the pronouns in this paragraph is 
conscious and purposeful, for in Western culture this kind of disem- 
bodied Promethean objectivity has been a masculinist preserve and 
privilege. Within its assumptions, woman has been conceived as part 


8 Sally Hacker, “The Eye of the Be- 
holder: An Essay on Technology and 
Eroticism,” in Doing it the Hard Way: 
Investigations of Gender and Technology, 
ed. (posthumously) Dorothy E. Smith 
and Susan M. Turner (Boston: Unwin 
Hyman, 1990), 214. 


79 These perspectives were pioneered 
by sociologists associated with the 
Edinburgh School as well as by feminist 
scholars Dorothy Merchant, Donna 
Haraway, Ruth Hubbard, Evelyn Fox 
Keller, Sandra Harding, and many 
others. See David Bloor, Knowledge and 
Social Imagery (London: Routledge & 
Kegan Paul, 1977); Donna Haraway, 
“A Manifesto for Cyborgs,” Socialist 
Review 80 (1985); Sandra Harding, The 
Science Question in Feminism (Ithaca, 
NY: Cornell University Press, 1986); 
Evelyn Fox Keller, Reflections on Gender 
and Science (New Haven, CT: Yale 
University Press, 1985); and Carolyn 
Merchant, The Death of Nature: Women, 
Ecology and The Scientific Revolution 
(New York: Harper & Row, 1985). 


3° Hofstadter, Gédel, Escher, Bach. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 21 


of nature, as “the sex,” and always embodied. In his Sixth Meditation, 
René Descartes provides the definitive articulation of this (masculine) 
stance when he asserts, “I am truly distinct from my body, and ... I 
can exist without it.”37 

Al’s positioning vis-a-vis the mind-body problem is shot through 
with contradiction. On the one hand, the Cartesian flight from em- 
bodiment and materialism reaches one of its clearest and most thor- 
ough articulations in the visionary statements of AIM, because the 
computer is “the embodiment of the world as the logician would like 
it to be,” not as it is.3? The goal of the AI scientist is to release mind 
from body: to download its contents into programs. On the other 
hand, however, AI and the new sciences of complexity are futuristic 
visions: “dreams of reason.” They are exercises of scientific imagina- 
tion rather than faithful codings of empirical reality. The dreamers 
know they are dreaming: They are not in denial about that. 

They simultaneously share and surrender the Cartesian dream of 
pure reason, of a “Promethean flight from embodiment,” to borrow 
Susan Bordo’s words.3? Their top-down struggle to release mind 
from body has, in the course of the history of AI research, paradoxi- 
cally pulled AI researchers back to biology. It is the body, the human 
biological system, with its brain, nervous system, nerve endings, and 
mercurial emotional apparatus, that weighs so heavily against Al 
and keeps its flight grounded. The more successful AI scientists are 
in advancing the Promethean dream, the more the model comes to 
resemble what they want to escape. From a bottoms-up perspective, 
Hofstadter describes the paradox that locks the AI scientist in a re- 
cursive loop: “[A]ll intelligences are just variations on a single theme: 
to create true intelligence, AI workers will just have to keep pushing 
to ever lower levels, closer and closer to brain mechanisms, if they 
wish their machines to attain the capacities which we have.”34 The 
better the bottoms-up machines get, the slower they will get. The mi- 
croworlds of the bottoms-up dream are the complete antithesis of the 
Buck Rogerian top-down supercomputer. 

The resulting discourse is, understandably, profoundly ambiva- 
lent about embodiment. The body is the enemy as well as the portal 
to knowledge that can transcend the body. In the Buck Rogers ver- 
sions of the dream, the program becomes the spaceship that allows 
the AI scientist-astronaut to escape from the enemy (e.g., the body, 
woman, morality, or nuclear annihilation). The scientist moves into 
another dimension, no longer human or embodied. The best of what 
he has to offer survives in this new dimension. In Hofstadter’s sce- 
nario, however, the scientist assumes a Zen-like stance and learns 
to live with, even savor, the intellectual and aesthetic pleasures of 
contradiction. He faces the paradox of the recursive loop head-on 


3* For a discussion of the Cartesian 
principle in relation to the sciences of 
complexity, regarding the bottoms-up 
approach, see Pagels, The Dreams of 
Reason. 


Bolter, Turing’s Man, 73. 


33 Susan Bordo, The Flight to Objectivity: 
Essays on Cartesianism and Culture 
(Albany: State University of New York 
Press, 1987). 


34 Hofstadter, Gédel, Escher, Bach, 579. 


22 WHAT WAS ARTIFICIAL INTELLIGENCE? 


and demonstrates the intellectual and aesthetic pleasures of life lived 

on its rim.3> 35 Hofstadter, Gédel, Escher, Bach. 
Whether Hofstadter’s version marks the end of AIM, the point 

where it transcends itself and mutates into the new sciences of com- 

plexity, or whether it marks Al’s rebirth as a mature research pro- 

gram, may still be an open question. He describes the top-down 

approach as over—“Retrospects”—and the bottoms-up perspective in 

the future tense—“Prospects.” These prospects are based on wholism 

rather than on reductionism, in computerese, parallel processing 

units and neural nets: “[MJany trains moving simultaneously down 

many parallel and crisscrossing tracks, their cars being pushed and 

pulled, attached and detached, switched from track to track by a 

myriad neural shunting-engines.”3° 3° Hofstadter, Godel, Escher, Bach, 623. 


THE POETRY IN THE PARADIGMS 


Like Descartes and Boyle, AI researchers embrace mechanical metaphors 

of mind. They conceive of mind as machine: a computer, a grid of 

electrical relay switches, “many trains moving simultaneously,” and 

so on. While some, like Turing, acknowledge the limitations of this 

conception within AI parascientific discourse and top-down AI mod- 

eling, the “program” is a metonymic surrogate for intelligence. Al 

constructs computer models of operations of mind by reducing its 

cognitive and biological processes to machine-recognizable inputs. 

AI modelers assume that all interesting manifestations of intelligence 

can be “captured” and “contained within” programs.37 According to 37 Minsky, Society of Mind. 

one journalistic chronicler of AI, some AI modelers even believe it is 

possible to precisely quantify and program the “odd little chemical 

electrical cloud of activity that is our personality.”3° 38 Fjermedal, The Tomorrow Makers, 7. 
Metaphors are usually thought of as tools of humanists, not scien- 

tists. The eighteenth-century English poet William Wordsworth was 

apparently the first thinker to argue that poets and scientists share 

similar relationships to nature, even though their languages differ. 

The scientist uses Royal Academy prose and the poet uses meter to 


interpret nature.3? Both approach the unknown through the por- 3° Hugh Kenner, The Mechanic Muse 
tals of the known, and scientist, no less than poet, uses analogies to i York: Oxford University Press, 
1987). 


construct bridges between the two. Although scientists from Fran- 
cis Bacon to the present wish it were not so, the bridges the scientist 
builds between the familiar and the mysterious, like the poet’s spans, 
are constructed of bricks baked in the cultural and linguistic kilns of 
historical time. 

Scientific vision, like poetic vision, is expressed most palpably 
through metaphors. The metaphors used by scientists are not, how- 
ever, incidental to the scientific enterprise. To the contrary, they em- 


WHAT WAS ARTIFICIAL INTELLIGENCE? 23 


power scientific vision; they provide the scaffolding for arguments, 
color the language of assertion, and guide inquiry. Indeed, Richard 
Dawkins claims, “Skill in wielding metaphors and symbols is one 
of the hallmarks of scientific genius.”4° In short, they are the magic 
carpets that make science possible.4" 

Metaphors are not, however, all that make science possible. Math- 
ematics formalizes and refines scientific vision; instrumentation am- 
plifies and standardizes it; and systematic, repetitive, and controlled 
observation tests its reliability. Yes, metaphors lurk within and enable 
these practices too—like Gédel’s Theorem, reminding us of the limits 
of all human knowledge. But lost at sea, who amongst us would not 
rather have a compass than a sonnet? Science has demonstrated its 
potency in practice. Studies in the history, philosophy, and sociology 
of science have nonetheless firmly established that metaphors are a 
necessary, though far from sufficient, component of scientific thought. 
Bacon was right! They are also mischief-makers that smuggle “the 
idols of the tribe” into science.4* This mischief does not negate or 
invalidate scientific claims, but it does humanize them. 

Social constructivist unpackings of the poetry in the paradigms 
have knit many scientific brows into exasperated consternation. But 
mythology! What are scientists to make of it? Hofstadter would 
probably advise art, music, or some more science; and he would 
be right. Mythology is a testament to human aspirations, not just a 
graveyard of human fallacies and foolishness. We are all, in some 
sense, poets, although there are very few Wordsworths, Shakespeares, 
Byrons, Bacons, Turings, or von Neumanns among us.*? All of our 
poetry, including the poetry of science is, however, a record of what 
humans value, aspire to, and fear. 

The Enlightenment cast scientists in the role of supermen. In its 
cosmology, nature displaced God as first principle; the scientist re- 
placed the priest as authoritative interpreter of the reality. Scientists 
were expected to see with the eyes of gods and to be nature’s ventril- 
oquists. The voice of scientific reason was conceived—impossibly—as 
the unmediated and therefore objective voice of nature. Scientific 
instrumentation and calculations created and preserved this construc- 
tion of objectivity, which did, in fact, prove to be an extraordinarily 
productive way of interpreting and imputing patterns to nature. In 
short, the Kantian trick usually worked.*4 

In making their daring claims at the height of the Inquisition, 
the members of the Royal Society not only risked the wrath of the 
God (if they were wrong) but the swords of inquisitors (if they were 
right). To weather the fury of the storm, fear was repressed in the 
tough-minded, even macho, Baconian vision of the masculine future 
of science. It was a brave vision that took modem science far. Like 


4 Richard Dawkins, Unweaving the 
Rainbow: Science, Delusion, and the 
Appetite for Wonder (Boston: Houghton 
Mifflin, 1998), 186. 

+ Mary Hesse, Models and Analogies 

in Science (South Bend, IN: University 
of Notre Dame Press, 1966); Bloor, 
Knowledge and Social Imagery; and 
Richard Rorty, “The Contingency of 
Language,” London Review of Books 17, 
April 17, 1986. 


# Bacon recognized that language 
contaminated the purity of science and 
longed for a purely scientific language, 
which would be culture-free. Most 
practicing scientists have, however, 
bracketed the problem of language and 
reported their results as if language 
were a neutral instrument. 


* George Lakoff has exhaustively 
explored the role metaphor plays in 
making ordinary language and thought 
possible. See George Lakoff and Mark 
Turner, More than Cool Reason: A Field 
Guide to Poetic Metaphor (Chicago: Uni- 
versity of Chicago Press, 1989). See 
also George Lakoff and Mark Johnson, 
Metaphors We Live By (Chicago: Uni- 
versity of Chicago Press, 1981); George 
Lakoff and Mark Johnson, Philosophy 

in the Flesh: The Embodied Mind and Its 
Challenge to Western Thought (New York: 
Basic Books, 1999); and George Lakoff, 
Women, Fire, and Dangerous Things: 
What Categories Reveal about the Mind 
(Chicago: University of Chicago Press, 
1981). 

#4 That is, at least at the level of em- 
pirical science, there appeared to be 
consonance in the patterns scientists 
discovered in or imputed to nature and 
the patterns perceived or constructed by 
the human mind. 


24 WHAT WAS ARTIFICIAL INTELLIGENCE? 


most brave visions, however, its monological and monovocal struc- 
ture and resonance left their imprimatur on both the vision and the 
visionaries. 

Fear is, of course, a proscribed emotion for men in the West (per- 
haps for men everywhere); if they have it, they are supposed to re- 
press it. Repressed sentiments and ideas do, however, have a habit 
of returning; mythology is a primary staging ground for this return. 
Male fear seems almost to be the axis upon which modern science 
has turned; and the momentum generated by this axis has been si- 
multaneously constructive and destructive to the species and the 
planet. For example, scientists did not seek to understand natural 
disasters just because they posed interesting scientific problems. 
Well-warranted fear, as much or more than cool-headed rationality, 


provided the momentum for the quest for scientific predictability and 


control. Earthquakes, hurricanes, volcanic eruptions, fires, floods, 
deadly diseases, nuclear explosions, and, yes, women have variously 
terrified many scientists as well as fascinated them. Not surprisingly, 
both terror and fascination are encoded in the mythopoetics of scien- 
tific thought. 


METAPHORS AND MINDS 


Metaphors based upon images of sexual relations and reproduction 
are both common and deeply embedded in the discourses of West- 
ern science and culture.45 Bacon himself incorporated them in the 
foundation documents of modern science, including his fragmentary 
The Masculine Birth of Time (1602 or 1603).4° These metaphors place 
the scientist in a hierarchical relationship of domination and control 
of nature.47 Within the mythopoetics of computer scientists, repro- 
ductive metaphors occupy a much more prominent position than 
copulative imagery, although the latter are invoked in predictable 
ways to represent inputs, circuitry, and connections. 

Images of male birthing have been a common motif (even, for 
those so inclined, a Jungian archetype) of Western origin stories—for 
example, Zeus giving birth to Athene from his head and God creat- 
ing Eve from Adam’s rib. Lionel Tiger claims rites of male bonding 
are “the male equivalent of child reproduction, which is related to 
work, defense, politics, and perhaps even the violent mastery and 
destruction of others.”4° Brian Easlea makes a similar point when he 
asserts: 


Men in prescientific societies, it may be generally argued, attempt to 
affirm masculine and, for them therefore, dominant status through 
secret exclusively male rituals. Quite often these rituals have a very 
direct “pregnant phallus” aspect to them, the male participants thereby 


45 Such understandings of generative 
metaphors are shared by nonfeminists 
as well as feminists. See, for example, 
Arthur Koestler, The Act of Creation 
(New York: Macmillan, 1967); Simone 
de Beauvoir, The Second Sex (New 
York: Vintage Books, 1974); Susan 
Stanford Friedman, “Creativity and the 
Childbirth Metaphor,” Feminist Studies 
13, no. 1 (1987); and Evelyn Fox Keller, 
Reflections on Gender and Science. 


4° Merchant, The Death of Nature; and 
Keller, Reflections on Gender and Science. 


47 Merchant, The Death of Nature; Keller, 
Reflections on Gender and Science; Hard- 
ing, The Science Question in Feminism; 
and Theodore Roszak, The Gendered 
Atom: Reflections on the Sexual Psychology 
of Science (Berkeley, CA: Conari Press, 
1999). 

# Lionel Tiger, Men in Groups (New 
York: Random House, 1969). 


WHAT WAS ARTIFICIAL INTELLIGENCE? 25 


demonstrating that through their special phallic powers they, like 
women, are able to give birth.49 


Both Easlea and Evelyn Fox Keller demonstrate the continuing pres- 
ence of the images of the “pregnant phallus” in the mythopoetics of 
contemporary science.>° 

Birth is the primary (perhaps even primal) source of most of the 
poetry in the paradigms of computer science. Computers are the 
sites of the generative process. They are, in the words of AI scien- 
tists Roger Schank and Harold Abelson, “omnipotent”; Schank and 
Robert Abelson describe them as “god.”5! They are also incubators, 
(male) wombs that are conceived as mediums for generating new 
forms of life. According to David Gelernter, these incubators will 
soon produce “mirror worlds”: You will be able to look into a “genie 
bottle on your desk” and see “reality.” Computers will soon become 
“crystal balls, telescopes, stained glass windows—wine, poetry or 
whatever—things that make you see vividly.” They will put “the uni- 
verse in a shoebox.” Why? Because “[a] bottled institution cannot 
intimidate, confound or ignore its members; they dominate it” (empha- 
sis in original).5? 

The virility and reproductive prowess of computers is expressed 
through three interconnected sets of birth images: images represent- 
ing creativity, immortality, and progress. 


IMAGES OF CREATIVITY 


Much of the mythologizing of the computer science fraternity is 
conscious, intentional, and programmatic: It serves a community- 
building function in AIM. It makes the work and the sacrifices it 
requires—the deferred gratification of always becoming, never fully 
arriving—special, ordained, daring, and even godlike. In God and 
Golem, Inc. (1964), Norbert Wiener, widely referred to as the “fa- 
ther of cybernetics,” maintains that machines that learn, reproduce 
themselves, and coexist with men pose profound theological ques- 
tions. Wiener points out that if a contemporary of Francis Bacon had 
claimed to be able to make machines that could “learn to play games 
or that should propagate themselves,” he would surely have been 
burned by the Inquisition, “unless he could convince some great pa- 
tron that he could transmute the base metals into gold, as Rabbi Low 
of Prague, who claimed that his incantations blew breath into the 
Golem of clay, had persuaded the Emperor Rudolf.”53 

According to the folklore of the computer science subculture, 
Wiener, John von Neumann, Gerald Sussman, Marvin Minsky, and 
Joel Moses all claimed to be actual descendants of Rabbi Low, per- 


49 Brian Easlea, Fathering the Unthinkable 
(London: Pluto, 1983), 17. 


5° Easlea, Fathering the Unthinkable; and 
Keller, Reflections on Gender and Science. 


5* Roger C. Schank and Robert P. Abel- 
son, Scripts, Plans, Goals and Under- 
standings: An Inquiry into Human Knowl- 
edge Structures (Mahwah, NJ: Lawrence 
Erlbaum Associates, 1977). 


52 David Gelernter, Mirror Worlds or 
the Day Software Puts the Universe in 

a Shoebox . . . How it Will Happen and 
What it Will Mean (New York: Oxford 
University Press, 1991), 1. 


53 Norbert Wiener, God and Golem, 
Inc.: A Comment on Certain Points 
Where Cybernetics Impinges on Religion 
(Cambridge, MA: MIT Press, 1964), 
49-50. 


26 WHAT WAS ARTIFICIAL INTELLIGENCE? 


haps the first mortal man to be credited with creating life without 
using woman as a vessel.>+ Moreover, Low’s descendants believe they 
are carrying on the family tradition. By the mid-1980s, these latter- 
day alchemists maintained that they had already given birth to four 
generations of Golem. The labor pains they were then experiencing in 
their attempt to give birth to “the fifth generation” of computers were 
extraordinary because the “pregnant phallus” was more pregnant 
than usual.>> It was struggling to bring forth very special progeny: a 
superchild who will be able to reproduce itself without the agency of 
either man or woman.>° 

Some enthusiasts herald “neural nets” as this special progeny. 
Indeed, to cross the border from one genre of scientific vision to an- 
other, an episode of Star Trek featured conscious, intentional, and eth- 
ical neural nets contemplating the injustices of their human sires. The 
crossover from science to science fiction is a common one: Science 
feeds the imagination of science fiction writers, and many scientists 
feed off of science fiction. As Freeman Dyson puts it, “Science is my 
territory, but science fiction is the landscape of my dreams.”97 Sci- 
entist and science fiction writer share the same imaginative field and 
vocabularies of motive: They are both posed on the precipice of the 
possible and asking, “What if . . . ?” In Greek mythology, the lesser 
Greek god Prometheus incites the wrath of Zeus by giving fire to 
man. Contemporary Prometheans invert the flight trajectory: Their 
leaps of imagination are intended to make them godlike. They seek 
to transcend embodiment, biology, and gravity, and give birth to a 
new, superior species of ideational forms. 


IMAGES OF IMMORTALITY 


According to the fathers-to-be, this much-anticipated superchild may 
cut through the genetic coding of the universe and produce “the 
next step in human evolution.”5° Some computer scientists believe 
this generation of computers will possess the power to transform 
their fathers into “supermen.”>9 They claim this vaunted son of the 
computer god will allow them to download the contents of their own 
minds into programs and thereby achieve immortality. 

One proud papa, Hans Moravec, director of the Mobile Robot Lab- 
oratory at Carnegie Mellon University, maintains, “The things we 
are building are our children, the next generations. They’re carrying 
on all our abilities, only they’re doing it better.”°° In Mind Children: 
The Future of Robot and Human Intelligence (1988), Moravec acknow]- 
edges that today “our machines are still simple creations, requiring 
the parental care and hovering attention of any newborn, hardly 


vw 


worthy of the word ‘intelligent.’” Within the next century, however, 


54 Turkle, The Second Self. 


55 Easlea, Fathering the Unthinkable. 


5° Edward A. Feigenbaum and Pamela 
McCorduck, The Fifth Generation: Arti- 
ficial Intelligence and Japan's Challenge to 
the World (New York: New American 
Library, 1984). 


57 Freeman Dyson, Imagined Worlds 
(Cambridge, MA: Harvard University 
Press, 1997), 9- 


58 Hans Moravec, Mind Children: The 
Future of Robot and Human Intelligence 
(Cambridge, MA: Harvard University 
Press, 1988). 

59 Michael Hirsch, “Computers En- 
visioned as Successors to Humans,” 
Buffalo News, June 14, 1987, 16(E). 


® Hirsch, “Computers Envisioned as 
Successors,” 16(E). 


WHAT WAS ARTIFICIAL INTELLIGENCE? 27 


he promises “they will mature into entities as complex as ourselves, 
and eventually into something transcending everything we know— 
in whom we can take pride when they refer to themselves as our 
descendants.”°! 

The gender of these children is seldom in doubt. When references 
to AI or robotics are personified, male pronouns are typically used. 
Within the often too transparently Freudian imagery of the lore of 
infotech, however, software and software designs are sometimes 
personified as females—for example, Eliza and Linda. This practice 
departs from common, humanistically inspired conventions of tech- 
talk because technology is usually personified as female.® Andreas 
Huyssen attributes this practice to fear of autonomous technology: 
“As soon as the machine came to be perceived as a demonic, inexpli- 
cable threat and as the harbinger of chaos and destruction ... writers 
began to imagine the Maschinenmensch as woman.”°3 This move also 
has mythological precedence, as, for example, Pandora’s box. 

For top-down AI, the signs are changed: The prospect of au- 
tonomous technology is exciting, a source of wonder and daring 
defiance of Judeo-Christian understandings of life and death and 
of conventional American values like God, motherhood, and apple 
pie. Sometimes this defiance gives practitioners pause, leads to self- 
interrogations of the ethical implications of AI. Yet, self-interrogations 
of the godlike powers of mind-makers are also, by definition, cele- 
brations of those godlike powers, which separate the dilemmas of 
AI scientists from the problems of ordinary folks, who are still stuck 
back in an earlier stage of evolution. An exception to top-downers’ 
embrace of autonomous technology is, however, made in the case of 
computer viruses, which carry the regressive stigma of biological life 
and usually infect only (female) software. 

The telos of AI is autonomous technology. It is AI’s ticket to immor- 
tality. Marvin Sussman, for example, conceives of the mind children, 
produced by AI, as delivering their fathers to the threshold of life 
everlasting: “[T]he machine can last forever,” and “if it doesn’t last 
forever, you can always dump it out onto tape and make backups.”°4 
As we shall see, bottoms-up AI is having difficulty sustaining this 
optimism. The return of biology not only refills Pandora’s box; it also 
opens the door to Mary Shelley’s humanist and feminist nightmare 
of the deformed progeny of phallic pregnancies: the Frankenstein 
monster.°5 

The Cartesian disconnection of AI researchers that permits them 
to conflate mind and machine also allows them to conceive of biolog- 
ical death as a minor episode in the life cycle of a superman: “If you 
make a machine that contains the contents of your mind, then that 
machine is you.”°6 Indeed, within the mythos of AI modelers, bio- 


& Moravec, Mind Children, 1. 


© Judy Wajcman, Feminism Confronts 
Technology (University Park, PA: 
Pennsylvania State University Press, 
1991); and Andreas Huyssen, “The 
Vamp and the Machine: Fritz Lang’s 
Metropolis,” in After the Great Divide: 
Modernism, Mass Culture, Postmodernism, 
ed. Andreas Huyssen (Bloomington: 
Indiana University Press, 1986). 

° Huyssen, “The Vamp and the Ma- 
chine,” 70; see also Judith Halberstam, 
“Automating Gender: Postmodern 
Feminism in the Age of the Intelligent 
Machine,” Feminist Studies 17, no. 3 
(1991). 


4 Sussman quoted by Fjermedal, The 
Tomorrow Makers, 8. 


65 Mary Shelley, Frankenstein (New York: 
St. Martin’s Press, 2000, original 1818). 


° Sussman quoted by Fjermedal, The 
Tomorrow Makers, 8. 


28 WHAT WAS ARTIFICIAL INTELLIGENCE? 


logical man (as well as woman) becomes an obstacle to be conquered 
and rationalized. 

The contents of the mind cannot be downloaded into immortality 
until the information channels are cleaned up. For this reason, AI 
simulation requires modelers to subject cognitive processes to the 
Law of the Hammer, albeit reluctantly and only for the time being 
until more complex forms of modeling become possible. The AI 
modeler must reduce complex cognitive and biological processes to 
a series of discrete and univocal binary commands. Modeling even a 
very simple movement like raising the arm of a man to press a lever 
may require identifying, mapping, and simulating hundreds, even 
thousands, of cognitive and neurological messages. Add to this the 
fact that within biological man, these messages are often confounded 
by the “noise” of indecision, procrastination, memory, reflection, love, 
lust, and other sentiments, values, and intentions that appear to be 
irrelevant to the immediate task at hand. 

Cleaning up the information channels to create models that will 
program a robot to push a lever with the same cool efficiency, regard- 
less of whether the lever releases bombs or coffee cups, is therefore 
a genuine achievement of Cartesian logic. Faulting the AI modeler 
for preferring clean channels to cluttered ones is like faulting the 
plumber for preferring clean drains to clogged ones. Both find their 
efforts blocked by the waste products of biological man. The AI 
modeler’s dream of a clean machine is a dream of Cartesian tran- 
scendence, perhaps even redemption. But where Descartes wanted to 
control the noise of embodiment, AI researchers frequently express 
a desire to eliminate the body. The late Heinz Pagels, then-director 
of the New York Academy of Sciences, found serious humor in the 
Cartesian mind-body problem: the incompatibility of rationality and 
sexuality, in this instance male sexuality. He opens his survey of the 
sciences of complexity, The Dreams of Reason: The Computer and the 
Rise of the Sciences of Complexity (1988), with a quotation from Robert 
Hutchins: “When the penis goes up, reason goes out the window.”°7 
Computers, it seems, can eliminate this distraction. 

Rodney Brooks explains why he wants to eliminate “the wet 
stuff”—human bodies—from the equation: “We are sort of locked 
into our genetic structure. At the moment we might be able to tweak 
our genetic structure a little bit, but nothing severe.”©8 Brooks sees 
“an advantage to building robots out of silicon and stuff like that, 
because we know how to control that fabrication process pretty 
well,” whereas we have “trouble with” biology: “We can’t add more 
brain cells to us, but we can add more processors, more silicon, to 
a robot.”©9 In short, robots are easier to expand, repair, and control 
than their messy and unpredictable prototypes. 


°7 Hutchins quoted by Pagels, Dreams of 
Reason, 19. 


® Brooks quoted by Fjermedal, The 
Tomorrow Makers, 33. 


° Brooks quoted by Fjermedal, The 
Tomorrow Makers, 33. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 29 


Because the legend of the pregnant phallus requires the scientist to 
make love to himself—to give birth to a “sacred image” of himself—it 
encourages narcissism.”° Sherry Turkle reports the following con- 
versation between AI scientists. Don Norman says, “I have a dream 
to create my own robot. To give it my intelligence. To make it mine, 
my mind, to see myself in it. Ever since I was a kid.” Roger Schank 
responds, “So who doesn’t? I have always wanted to make a mind. 
Create something like that. It is the most exciting thing you could 
do. The most important thing anyone could do.” Gary Drescher tells 
Turkle, “We have the right to create life, but not the right to take our 
act lightly.””7" Drescher believes scientists have ethical obligations in a 
society where human and artificial intelligence live together. 

Following the lead of science fiction writer Isaac Asimov, Drescher 
entertains the idea that AI may make a new form of murder possible: 


People always talk about pulling the plug on computers as though 
when it comes to that they will be saving the world, performing the ul- 
timate moral act. But that is science fiction. In real life, it will probably 
be the other way around. We are going to be creating consciousness, 
creating lives, and then people may simply want to pull the plug when 
one of these intelligences doesn’t agree with them.7? 


IMAGES OF PROGRESS IN AI DISCOURSE 


Some AI scientists acknowledge that the next step in evolution may 
render humans obsolete. Marvin Minsky thinks “people will get fed 
up with bodies after a while.”73 He predicts that like the dinosaurs 
we might disappear, leaving behind a “society” of interacting and 
self-generating computer systems.74 

Evolutionary analogies are common in AI discourse. They appear 
to represent a form of masculine display: a way of saying my science 
is bigger (more potent or pregnant) than yours. However, evolution- 
ary images are also used to convey disdain for and distance from 
conventional conceptions of life, death, thought, and morality. That 
is, they are used to signal a radical departure from all previous ways 
of knowing and being in the world. Thus, Moravec asserts, “I have 
no loyalty to DNA,” and Mike Blackwell claims, “Bodies have served 
their purpose.”75 

Moravec valorizes the departure, the irrevocable break with the 
past: “We are on a threshold of a change in the universe comparable 
to the transition from non-life to life.”7© On one level, AI scientists 
seem to be embracing a return to pre-Baconian animism in which 
matter, cum machine, is endowed with life and anthropomorpho- 
sized. There is, however, more to the equation. The transition is not 
to life. There is a change in signs, which negates the value of human 


7 Donna Haraway, presentations and 
discussions at Science as Cultural 
Practice, a summer institute sponsored 
by the National Endowment in the 
Humanities, Wesleyan University, 
Middletown, CT, July 1991. 


7 Drescher quoted by Turkle, The Second 
Self, 261. 


7 Drescher quoted by Turkle, The Second 
Self, 262. 


73 Minsky quoted by Fjermedal, The 
Tomorrow Makers, 7. 


7 Minsky, Society of Mind. 


75 Moravec and Blackwell quoted by 
Fjermedal, The Tomorrow Makers, 60. 


7° Moravec quoted by Fjermedal, The 
Tomorrow Makers, 60. 


30 WHAT WAS ARTIFICIAL INTELLIGENCE? 


life: Machines evolve, humans download or die. Within AI’s mech- 
anistic reconstruction of evolutionary theory, the pregnant phallus 
finally achieves deliverance: Mind is released from body and man 
is released from his biological dependence on woman. Moravec de- 
scribes the brave new “post-biological” world of AT: 


All our culture can be taken over by robots. It'll be boring to be human. 
. .. We can’t beat the computers. So it opens another possibility. We 
can survive by moving over into their forms . . . because we exist in a 
competitive economy, because each increment in technology provides 
an advantage for the possessor. . . . Even if you can keep them [the 
machines] slaves for a long time, more and more decision-making will 
be passed over to them because of the competitiveness. 


We may still be left around, like the birds. It may be that we can ar- 
range things so the machines leave us alone. But sooner or later they'll 
accidentally step on us. They'll need the material of the earth.77 


REPRODUCTION AS DESTRUCTION IN AI 
DISCOURSE 


In the transition from life to program, the clean machine supersedes 
its sweaty, plodding, loving, lusting, and aging progenitor. And, the 
pregnant phallus eliminates the “wet stuff” that permitted its pro- 
totype to penetrate Baconian “holes and corners.”7° The violence 

of the vision is neatly occluded by comic strip captions. Robots will 
accidentally step on “us,” but that’s okay because “we” won’t really 
be there anyhow: “our” now immortal minds will be able to aban- 
don mother Earth entirely. Indeed, some AI scientists invoking the 


doomsday scenario believe it is imperative that “we” get some minds 


off of this nuclear and ecologically endangered planet and into space 
colonies before it is too late. 

Inevitably, the question must be raised: Which minds? Since the 
capacity of the most powerful parallel processing machines (connec- 
tion machines) will be finite, not everyone will be able to get out of 
their bodies or off of the planet. Some of “us” will be stepped on, 
incinerated, or poisoned by toxic waste. So, who gets downloaded 
into the programs? The new evolutionary logic dictates the answer. 
The best minds, of course, the kinds of minds that are most readily 
available for modeling in the AI laboratories at MIT, Stanford, and 
Carnegie Mellon University: minds of upper middle-class, white, 
American, predominantly male computer scientists. These are, not 


incidentally, some of the same minds and bodies that are most sought 


after by sperm bank entrepreneurs and their customers. 
These are also some of the same minds that envision a future 
in which AI will render participatory democracy obsolete. Among 


77 Moravec quoted by Hirsch, “Comput- 
ers Envisioned as Successors,” 16. 


7% Bacon quoted by Merchant, The Death 
of Nature, 168. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 31 


them are minds that herald the coming of a time when machines, 
not people, will control the world’s nuclear arsenals; when new 
forms of slavery will be introduced in which living machines (cy- 
borgs) programmed to be “ethical” will serve as slaves; when robots 
will be programmed to meet all (in- and out-of body) erotic needs 
and thereby render human intercourse and biological reproduction 
redundant.’ In short, these are minds that embrace what Neil Post- 
man calls “technopoly,” or totalitarian technocracy.°° 

The mythos and metaphors of Al talk and texts display a familiar 
design. AI discourse is a discourse of control; it builds hierarchy into 
the hard-wiring of its circuitry. The robotic fantasies of AI researchers 
presuppose the necessity of “the violent mastery and destruction 


of others.” 


Comic book talk papers over the perversity of AI con- 
cepts of creativity, immortality, and progress, but MIT researcher and 
outspoken in-house critic of AI ideology and eschatology, Joseph 
Weizenbaum, cuts through the cartoon images and conceives the per- 
versity within the same frame history has used to comprehend its 
previous incarnations: genocide.®? 

The faded mythology encoded in AI talk and texts demonstrates 
that AI is not the univocal discourse—not the pure Cartesian reason— 
that its architects thought they were encoding. Like the technostrate- 
gic discourse of the Cold War defense intellectuals analyzed by Carol 
Cohn, AI is also a discourse, which fails according to its own crite- 
ria: It is as far from a “paragon of cool headed rationality” as was 
Francis Bacon’s belief in the diabolical powers of witches.°3 Weizen- 
baum’s characterization of AI scientists as big children who have not 
given up their “sandbox fantasies” or sublimated their dreams of om- 
nipotence may be correct.54 But lest we swell with the satisfaction of 
one-upping would-be gods, we should remember that we all harbor 
lost children within us, and that fear can usually be counted upon to 
release them from captivity. And fear was the generative core of Cold 
War cosmology. 

Let us remind ourselves that the big children of AIM possess some 
of the best scientific and mathematical minds of the age. They are 
members of a powerful scientific elite: researchers, teachers, and gate- 
keepers of the most advanced and prestigious academic and com- 
mercial computer research centers in the world. The metaphors these 
men use to conceive nature, gender, and computer architectures are 
far more potent (and pregnant) than yours or mine. Donna Haraway 
contends that biology has already undergone a cybernetic revolution, 
in which natural objects have been retheorized as “technological de- 
vices properly understood in terms of mechanisms of production and 
storage of information.”*5 This metaphoric reconfiguration of the ter- 
ritory of science has fundamentally changed the character of scientific 


79 Fjermedal, The Tomorrow Makers. 
8° Neil Postman, Technopoly: The Surren- 


der of Culture to Technology (New York: 
Alfred A. Knopf, 1992). 


& Tiger, Men in Groups, 69. 


® Joseph Weizenbaum, Computer Power 
and Human Reason (San Francisco: W. H. 
Freeman, 1976). 


83 Carol Cohn, “Sex and Death in the 
Rational World of Defense Intellectu- 
als,” Signs 12, no. 4 (1987): 717. 


84 Joseph Weizenbaum, “Not Without 
Us,” Z Magazine, January 1988, 94. 


85 Donna Haraway, “The Biological 
Enterprise,” Radical History Review 20 
(1979): 223. 


32 WHAT WAS ARTIFICIAL INTELLIGENCE? 


interventions in the biological and material worlds, and has thereby 
changed the nature of those worlds. The generative metaphors of 
information processing have transformed humans into cyborgs and 
astronauts—all of us: technophiles and technophobes, feminists and 
misogynists, acrobats and apple growers too.%° 

Unlike Bacon’s patriarchal metaphors, which saw knowledge 
issuing from a chaste marriage between men’s mind and nature, 
top-down cybernetic metaphors locate the genesis of knowledge in 
the marriage of men’s minds and male machines. The mythos of 
male bonding encoded in AI discourse bears little resemblance to 
Plato’s homoerotic vision. AI metaphors replace Eros with objects: 
fetishes made of circuits and chips. Where Baconian epistemology 
suppressed the female principle, AIM’s technovisions negate the 
human principle, and as Weizenbaum points out, “There’s nothing 
left after you’ve destroyed the human species.”87 


Co1tTus INTERRUPTUS 


The strong top-down AI research program is both a tribute to and 
testimony against Western dualism and Enlightenment conceptions 
of reason. The self-correcting elements in the hyperrationality of 
the top-down AI program were powerful enough to discover AIM’s 
own limits. This discovery, in turn, invalidated the essential tenet 
of Al’s premise: that reason exists in a dimension apart from and 
beyond history, culture, and sentient beings. The failure of the top- 
down program was a triumph for biology: a regrounding of mind 
in body and of mental processes as human, learned, and socially 
situated. Promethean man was pulled back to earth, as he always is 
when he flies too high: too far from his origins. In the mythopoetics 
of Western dualism, the triumph of the body is a triumph of the 
feminine principle. 

In AI parascientific discourse, the latent symbolic ascent of the 
feminine that accompanied the paradigmatic shift to the bottoms-up 
approach was never grasped, and appears in any case to have been 
ephemeral: a transitory return of the repressed feminine dimension. 
It was briefly ascendant at the point of impact and (yes, I will say 
it) intercourse of thesis (top-down) and antithesis (bottoms-up), and 
in the period of the reconceptualization and rebirthing of research 
programs for AI that immediately followed. That this moment of 
opening occurred at the same time that the larger social, cultural, and 
political formations of global power were also undergoing profound, 
even epochal, transformations is, as we have seen, not coinciden- 
tal. The crises that the end of the Cold War posed for the defense 
industry and for research funded by the Defense Department were 


86 Donna Haraway, “A Manifesto for 
Cyborgs”; and Romanyshyn, Technology 
as Symptom and Dream. 


87 Weizenbaum quoted in Fjermedal, The 
Tomorrow Makers, 140. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 33 


widely chronicled in the media in the late 1980s and early 1990s. The 
permanent war economy had been very costly, but it had insulated 

postwar America against the extremes of the boom and bust cycles 

of capitalism. What was at the time dubbed by The New York Times as 
“risks of peace” included not only displacing the economic stabiliza- 
tion of defense spending but also displacing defense workers, which 
included a highly educated techno-scientific strata that could be very 


dangerous if it became alienated.8° 88 See Sue Curry Jansen, “When the 
Center No Longer Holds: Rupture 

. and Repair,” in Critical Communica- 
launched what President George Bush called a “new world order,” tion Theory: Power, Media, Gender, and 


which, counterfactually, sought to keep the old power-knowledge Technology (Lanham, MD: Rowman & 
Littlefield, 2002), for a fuller discussion 
of news-framing at the end of the Cold 
as television, and it was not, in any case, a strategy that held much War. 


The dramaturgical accompaniments of the Persian Gulf War 


of the military-industrial complex intact. It did not work, except 


long-term promise: It was too expensive and morally repellent. For 
example, the US military estimates that the brief war took some- 
where between one hundred thousand and two hundred thousand 
Iraqi lives, most of them civilians. As a dazzling, well-edited, globally 
broadcast television display of the triumphs of American techno- 
culture, it did, however, foreshadow the future. The Clinton-Gore 
Administration defined that future in its technovision, the National 
Information Infrastructure, and in its policies, treaties, and legislative 
initiatives (NAFTA, GATT, and the omnibus US Telecommunications 
Act of 1996), which supported the creation of a US-dominated global 
information economy. Clinton-Gore went where no Republicans 
could have dared to go in accelerating the growth of corporate power 


and in defining corporate “competitiveness” as a defense initiative.59 %9 Department of Defense (web- 
site), accessed March 2000, 


The “smart” bombs profiled in the Persian Gulf War drama were et 
https: / /www.defenselink.mil. 


prototypes for the smart technologies that would build the new in- 
formation economy. By the early 1990s, the bombs were almost smart 
enough, and the research that produced them had already had some 
success in the consumer marketplace—for example, the original com- 
puter game, Flight Simulator, and search-and-destroy video games. 
Virtual reality simulations showed commercial promise as techno- 
entertainments as well. AI research, like other forms of defense re- 
search, was encouraged to redefine itself, and generous government 
funding was dedicated to moving American science, scientists, and 
defense contractors through the transitional period. Research agen- 
das were expanded to include educational, entertainment, biotech- 
nologies, and other commercial applications. Visionary high-tech 
ideas were brought to bear on mundane tasks. Military and com- 
mercial agendas were often pursued in tandem. The development 

of robotic vision, for example, retains military applications, making 
those smart bombs even smarter in hitting targets, but its potential 
applications as prosthetics for the blind are also smart commercial 
(and humane) investments. 


34 WHAT WAS ARTIFICIAL INTELLIGENCE? 


This commercialized technovision appears to support somewhat 
more humane agendas than the mature top-down AI approach, in- 
sofar as it is less overtly tied to the monovocal agenda of Cold War 
demonology—for example, eradicating the “evil empire.” The level 
of fear and doomsday paranoia that accompanied the Cold War vi- 
sion had largely disappeared from mass-mediated articulations of 
ideology and public policy until the 2001 terrorist attacks on the US 
and the US’s subsequent launching of its global War on Terrorism. 
This demonology continued to thrive in defense think tanks, and it 
prospers among fringe militia, survivalist, and white supremacist 
groups: groups made up primarily of white males who claim to be 
disenfranchised by the moderately more inclusive post-Cold War 
definitions of social reality. 

Within computer science, doomsday scenarios have, interestingly 
enough, been transferred to the programs themselves. They revolve 
around fears of techno-terrorism, including hacker breaches of gov- 
ernment and corporate security, scenarios of contamination of net- 
worked systems by massive self-replicating viruses, and dystopias 
involving techno-wars and extermination of the human race by a 
future species of intelligent robots. 


WILL THE PREGNANT PHALLUS DELIVER SELF- 
REPLICATING POWERS AND A NEW GENERATION 
OF ANTI-HUMAN TERRORS? 


What does the future hold? That is the perennial question that is 
posed to, and by, AI and robotics research and development scien- 
tists. The AI research community is no longer fully a male preserve, 
“a world without women.” Women are a growing, though still small, 
presence within the ranks of AI and robotics research and develop- 
ment, although they have not yet issued any manifestos. Whether 
they will ultimately forge new metaphors and new ways of thinking 
about conceiving artificial life remains an open question. 

At present, the US government under President George W. Bush 
is gearing up once more to strongly reassert its presence in computer 
science research and development by reviving development of the 
ill-fated (and, many scientists believe, ill-conceived) Cold War Star 
Wars missile defense system, originally proposed and funded under 
the Reagan-Bush I Administration.” It appears, at this point, that 
the War on Terrorism has given the Bush Administration the man- 
date it needs to override scientific reservations and congressional 
opposition to reviving the missile defense program. Moreover, the US 
government has announced that it now needs aggressive as well as 


%° For a critical analysis and summary of 
the scientific and political debates about 
the Star Wars program, see Frances 
Fitzgerald, Way Out There in the Blue: 
Reagan, Star Wars, and the End of the 

Cold War (New York: Simon & Schuster, 
2000). 


WHAT WAS ARTIFICIAL INTELLIGENCE? 35 


defensive weapons to conduct cyber-warfare. The huge reinfusion of 
defense funds will define the futures of AI and AL. 

Will the mature research program of the bottoms-up approach 
of AI be more humane than the mature program of the top-down 
approach? There is no reason to assume it will be; indeed, it could 
be more inhumane. The sandbox fantasies have not disappeared; in 
fact, they may have moved closer to becoming technological reali- 
ties. Hans Moravec is still around, and still believes that “biological 
humans” will “be squeezed out of existence.”9* Danny Hillis, now 
known as the father of parallel processing, is still thinking about es- 
caping the grim reaper: “I’m as fond of my body as anyone, but if 
I can be 200 with a body of silicon, I’ll take it.”9* Ray Kurzweil is 
predicting we will become robots or fuse with robots. 

In “Why the Future Doesn’t Need Us” (Wired, March 2000), Bill 
Joy, cofounder and chief scientist of Sun Microsystems and co-chair 
of a presidential commission on the future of information technology 
research, wonders how other techno-wizards can silently live with 
their fears. Joy reports that the kind of technology Moravec envisions 
will be feasible by 2030: 


What was different in the 20th century? Certainly, the technologies 
underlying the weapons of mass destruction (WMD)—nuclear, bi- 
ological, and chemical (NBC)—were powerful, and the weapons an 
enormous threat. But building nuclear weapons required, at least for a 
time, access to both rare—indeed, effectively unavailable—raw materi- 
als and highly protected information; biological and chemical weapons 
programs also tended to require large-scale activities. 


The 21st-century technologies—genetics, nanotechnology, and robotics 
(GNR)—are so powerful that they can spawn whole new classes of 
accidents and abuses. Most dangerously, for the first time, these acci- 
dents and abuses are widely within the reach of individuals or small 
groups. They will not require large facilities or rare raw materials. 
Knowledge alone will enable the use of them. 


Thus we have the possibility not just of weapons of mass destruction 
but of knowledge-enabled mass destruction (KMD), this destructive- 
ness hugely amplified by the power of self-replication. 

I think it is no exaggeration to say we are on the cusp of the fur- 
ther perfection of extreme evil, an evil whose possibility spreads well 
beyond that which weapons of mass destruction bequeathed to the 
nation-states, on to a surprisingly terrible empowerment of extreme 
individuals. 

Where the NBC technologies of the twentieth century were largely 
developed by the military in government-controlled laboratories, 

Joy points out, “We are aggressively pursuing the promises of these 
new technologies within the now-unchallenged system of global 
capitalism and its manifold financial incentives and competitive 


% Joy, “Why the Future Doesn’t Need 
Us,” 2. 


% Joy, “Why the Future Doesn’t Need 
Us,” 2. For a recent account, which 
demonstrates that Al’s self-promoting 
hype continues despite claims of a 
new humility within AI and AL, see 
Jim Krane, “ ‘Human’ Robots March 
Forward—on Movie Screen and Off,” 
Morning Call (Allentown, PA), June 26, 
2001, 8(D). 

3 Joy, “Why the Future Doesn’t Need 
Us,” 2. 


94 Joy, “Why the Future Doesn’t Need 
Us,” 3. 


36 WHAT WAS ARTIFICIAL INTELLIGENCE? 


pressures.”95 He envisions scenarios where corporations may be 
forced into something like voluntary disarmament or the equivalent 
of biological weapons inspections if the species is to survive. Lest 

we blame the messenger, Joy is pushing the panic button precisely 
because he wants to initiate public dialogue about techno-futures, 
which, to date, have been shaped without it. He sees the astronautic 
fantasies of scientists, which call for evacuating the earth, as forms of 
denial that abdicate responsibility. 

In the aftermath of the anthrax attacks on the US Postal Service, 
Congress, the media, and the public that immediately followed the 
September 11, 2001 terrorist attacks in New York, Washington, and 
Pennsylvania, Joy’s jeremiad resonates with even greater gravity. 

No one has taken responsibility for the anthrax attacks, which are 
at present assumed to be the work of a domestic terrorist, not the 
al-Qaida network. Government forces, from federal to local levels, 
remain on high alert for further biological and chemical terrorist 
attacks. As a result, the scenario Joy describes takes on a new and 
chilling sense of reality. 

Joy does not address the gendered components of these technovi- 
sions or the gender orders they will support, but he does try to see 
beyond the conventional horizons of Western science and culture. 
Siding with the biological life on planet Earth, he is a de facto ally 
in the struggle for a more human, and therefore a future friendlier 
to women, the species, and the planet. Moreover, by virtue of the 
authority his background gives to his argument, it is a valuable addi- 
tion to the arsenal of ideas that can be mobilized in “the semiologi- 
cal warfare” that is required to interrupt the privatization of policy 
making that the new enclosure movement has empowered and nor- 
malized.% Joy’s goal is to open up a broadly based dialogue about 
the deployment of technologies before they are deployed, not to pro- 
vide lay readers with a definitive take on the science of the future. 

If his scenario is alarmist, then open, informed, critical, democratic 
dialogue can serve as a corrective. In any case, democratic dialogue 
about techno-futures is urgently needed if democracy is to retain (or 
recover) any meaning beyond the symbolic or spectral.97 


NEw SCIENCE REQUIRES NEW POETRY: 
RETURNING TO THE LAUNCH PAD 


The process of interrupting and correcting the talk and texts of 
technoscience has just begun. Haraway describes such interven- 
tions as forms of practicing “politics by other means.”9® Yes, and 
poetry too! For the first step in scientific revolutions (as in political 


95 Joy, “Why the Future Doesn’t Need 
Us,” 7. 


9 Umberto Eco, Travels in Hyper-Reality 
(New York: Harcourt Brace Jovanovich, 
1986). 


97 Guy Debord, The Society of Spectacle 
(New York: Zone Books, 1995). 


% Donna Haraway, “Primatology is 
Politics by Other Means,” in Feminist 
Approaches to Science, ed. Ruth Bleier 
(New York: Pergamon Press, 1986), 
77-118. 


WHAT WAS ARTIFICIAL INTELLIGENCE? 37 


revolutions) is to change the names, because scientific revolutions 
are metaphoric redescriptions of nature, not (or not only) codings of 
revolutionary new insights into the intrinsic nature of phenomena. 

Most, though certainly not all, constructivist and Western femi- 
nist conceptions of nature and humankind break with the astronau- 
tic vision. Feminist perspectives support metaphors, models, and 
taxonomies for describing nature that are less hierarchical, more 
contextual and permeable, and perhaps more reflexive than their 
masculinist predecessors. The achievements (and perversions) of 
Western science have been possible because they have taken flight 
on transcendent metaphors—for example, Prometheus, Icarus, Faust, 
Superman, cyborg, and astronaut.'°° These metaphors deny embod- 
iment and mortality, whereas images of domesticity, embodiment, 
and material necessity—images drawn from women’s experience— 
keep our feet on the ground. Fully human conceptions of nature and 
being must do both or both/and.*°" That is, they must allow us— 
all of us!—to dream dreams that make the impossible possible. But 
they must also recognize that it takes many dreamers—many diverse, 
self-reflexive, human agents—to dream life-affirming dreams: women 
and men in life-sustaining communities, not insular enclaves of scien- 
tific geniuses or self-replicating forms of hardware and software. 

This new way of thinking is, however, unlikely to emerge from 
truces, whether voluntary or mandated, in the so-called scientific 
wars. Adding women to science and stirring will not do the job. In- 
deed such a strategy, no matter how well intended, is likely to either 
kill the spirit of the women who are added to science or kill the spirit 
of science. Rather, species-friendly conceptions of nature are far more 
likely to find incubation within new generative metaphors that will, 
in turn, prove to be more illuminating, inspiring, and effective in 
meeting the life-sustaining challenges that lie ahead. Or, to put it 
pragmatically, expanding the landscape of the scientific imagination 
may prove to be more important to twenty-first-century earth science 
than it was to twentieth-century space science. 

If scientists like Moravec, Kurzweil, Joy, and others are right about 
the future, survival of the planet now requires terminating the ex- 
terminating elements in the self-replicating technologies of genetics, 


102 Our interventions need to attend to 


nanotechnology, and robotics. 
the problem from the launch pad rather than the space station. We 
need to cast our collective lot with Earth, not the stars. We need to 
find our metaphors closer to home: to come back to Earth, back to 
our aging, sweating, imperfect, mortal bodies. We need to face the 
responsibilities, tensions, ambiguities, and pleasures of a fully human 
life and death. In short, we need to dream a new cultural dream: a 
dream that requires nothing less than interruption and redirection of 


the out-of-body experiences of modern and postmodern science. 


99 Mary Hesse, Revolutions and Recon- 
structions in the Philosophy of Science 
(Bloomington: Indiana University Press, 
1980); and Rorty, “The Contingency of 
Language.” 


7°° See discussion in Sue Curry Jansen, 
“Introduction: Scholarly Writing Is 

an Unnatural Act,” in Critical Commu- 
nication Theory: Power, Media, Gender, 
and Technology (Lanham, MD: Row- 
man & Littlefield, 2002), as well as Eve 
Tavor Bannet, “The Feminist Logic of 
Both/And,” Genders 15 (1992). 

*°t Bordo, The Flight to Objectivity; Clive 
Hart, Images of Flight (Berkeley: Uni- 
versity of California Press, 1988); and 
Romanyshyn, Technology as Symptom and 
Dream. 


7° Joy, “Why the Future Doesn’t Need 
Us.”